AI Chatbots for University Student Mental Health: Bibliometric Mapping and Systematic Review with Future Directions for Sustainable Development Goals

AI Chatbots for University Student Mental Health: Bibliometric Mapping and Systematic Review with Future Directions for Sustainable Development Goals | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Systematic Review AI Chatbots for University Student Mental Health: Bibliometric Mapping and Systematic Review with Future Directions for Sustainable Development Goals Xing Yi, Ying Chen, Sau Cheong Loh, Yan-Li Siaw, Li Jie Hao, Hassan Abuhassna This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7486943/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 13 You are reading this latest preprint version Abstract University students worldwide are experiencing increasing mental health challenges, while existing support resources remain insufficient. Artificial intelligence–based chatbots have emerged as scalable and accessible tools for psychological support, aligning with the Sustainable Development Goals on health, quality education, and well-being. This study aims to map global research trends on AI-based chatbots for mental health in higher education and to identify future research directions. A bibliometric analysis was conducted using 58 English-language articles retrieved from Web of Science and Scopus. Publication patterns, influential journals, leading countries, and keyword networks were analyzed. Additionally, a systematic literature review of 20 selected studies was performed to synthesize research gaps and thematic priorities. The bibliometric analysis revealed a sharp increase in publications since 2020, with research concentrated in a limited number of countries and journals. Thematic synthesis highlighted five priority areas: stress management, mental health symptoms, emotional support, intervention design, and user interaction. However, the review also identified limited diversity in research designs and narrow applications of chatbot technologies. Findings demonstrate growing scholarly interest in AI-based chatbots for student well-being, yet significant gaps remain. Future research should focus on inclusive and user-centered chatbot design, integration of evidence-based interventions, international cooperation and cross-cultural validation, and ethical issues. These insights advance both theoretical understanding and practical development of AI-based mental health tools, contributing to the achievement of relevant Sustainable Development Goals. Biological sciences/Psychology Social science/Psychology Scientific community and society/Scientific community Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 1. Introduction University students are increasingly experiencing psychological distress due to academic demands, career uncertainty, and social challenges (Karyotaki et al. 2020 ; Mofatteh 2021 ). As they represent the next generation of societal leaders, safeguarding their mental health and well-being is both an educational priority and a broader public health concern. It not only supports their academic success but also contributes to global aspirations for healthier, more inclusive, and equitable societies, as reflected in international frameworks in United Nations’ 2030 Agenda for Sustainable Development Goals (SDG), such as SDG 3 for Good Health and Well-being, SDG 4 for Quality Education, and SDG 10 for Reduced Inequalities (Nations 2015 ). Although universities have introduced initiatives such as mental health courses and counseling services, these resources are often overwhelmed, difficult to scale, and lack personalization (Priestley et al. 2022 ). This creates disparities in access to timely support, particularly for students from underserved backgrounds. Addressing these gaps requires innovative, scalable solutions that can complement traditional services while maintaining quality and equity. In recent years, AI-based chatbots have emerged as promising tools for mental health intervention, offering continuous availability, cost-effectiveness, and user anonymity. Advances in generative AI (e.g., GPT-3, GPT-4) have enabled chatbots to provide empathetic, context-aware dialogue (Bansal et al. 2024 ), making them attractive complements to traditional counseling, especially for university student populations (H. Liu et al. 2022 ; Manole et al. 2024 ). While existing studies have explored chatbot effectiveness for various mental health outcomes, they often focus on specific applications or short-term results. There remains a lack of comprehensive analysis that maps the intellectual structure of this field, identify influential contributors, and reveal thematic trends. Long-term considerations, such as ethical risks and sustainable development, also require further attention. Therefore, this study explicitly positions itself within the pursuit of the SDGs and integrates bibliometric methods with a systematic literature review. The research objectives are to chart publication trends, identify key contributors (journals, countries, institutions, landmark papers), trace thematic evolution over time, and uncover the future directions. The specific research questions are: 1.Publication trends: How have publications on AI chatbots for university student mental health been distributed over time? 2.Key contributors: Which journals, countries, institutions, and papers have been most influential in this field? 3.Thematic trends: What are the key words, prevailing research topics, and how have they evolved? 4.Future directions: What are suggested directions for future research in AI-based student mental health support related to SDG? 2. Literature Review 2.1 AI-based chatbots and mental health potentials AI-powered chatbots are computer programs that use natural language processing to simulate human conversation. In recent years, particularly transformer-based generative models (GPT-3, GPT-4), have greatly enhanced chatbots’ abilities to mimic human empathy and responsiveness (Adamopoulou and Moussiades 2020; Bansal et al. 2024 ). These advanced models produce adaptive, nuanced replies, improving the quality of conversational support (Bansal et al. 2024 ). By leveraging sentiment analysis, chatbots can recognize users’ emotion and respond with empathic understanding, which can facilitate emotional disclosure and reduce perceived stigma (Denecke and Gabarron 2024 ). This reduction in stigma is crucial for achieving equitable access to mental health support, which is aligned with the aim of SDG 10 to reduce various forms of inequality (Nations 2015 ). From a psychological standpoint, such empathic engagement aligns with social support theory (Jones 1985 ): when users feel heard and understood, stress and loneliness decrease (Syed 2024 ). Chatbots can also deliver structured interventions, such as cognitive behavioral therapy (CBT) or exercises and guided self-reflection (Eltahawy et al. 2024 ; Farzan et al. 2025 ). Their potential to offer scalable, basic mental health support significantly contributes to Sustainable Development Goal 3 and 4. This target aims to "reduce by one third premature mortality from non-communicable diseases through prevention and treatment and promote mental health and well-being" by 2030 (Nations 2015 ). Despite this potential, most research to date has focused on the technical development of chatbots or isolated case studies. Few studies integrate technical and psychological perspectives, especially within the higher education context. In other words, literature often examines chatbot algorithms or short-term efficacy, without synthesizing broader trends or combining insights from computer science and mental health disciplines. 2.2 AI-based chatbot in supporting university students’ mental health University students are generally considered a high-risk group for stress and mental illness (Karyotaki et al. 2020 ). To support student well-being, many higher education institutes offer mental health courses and free counseling, but demand often exceeds capacity (Hyseni Duraku et al. 2023). In this setting, chatbots have significantly expanded access to mental health resources. They provide available, scalable, and low-cost support, which can lower barriers for students facing financial or stigma-related obstacles (Manole et al. 2024 ; Omarov et al. 2023 ). Consequently, chatbots have become a popular option for students seeking help (H. Liu et al. 2022 ). By offering immediate guidance and coping strategies, these tools can contribute to early intervention, emotional regulation, and overall well-being (Klos et al. 2021 ; H. Liu et al. 2022 ). AI chatbot contributes to their academic success and overall well-being, core aspects of SDG 4 "ensure inclusive and equitable quality education"(Nations 2015 ). Recent studies highlight the promise of generative AI chatbots for student mental health (Klos et al. 2021 ; H. Liu et al. 2022 ). However, most focus on singular issues (e.g., reducing anxiety or stress) without providing a holistic view of student mental health. Existing meta-analyses and reviews have confirmed chatbots’ effectiveness for youth mental health (Feng et al. 2025 ; H. Li et al. 2023 ) or categorized chatbot interventions by type and approach (Omarov et al. 2023 ). Yet many of these analyses do not specifically address the higher education context. Students’ psychological needs are often more complex and multifaceted, influenced by academic pressures and transitional life stages. Previous literature lacks a broad mapping of the field of AI chatbots for university student mental health. Several gaps remain. First, there is a lack of bibliometric analyses that systematically uncover the intellectual structure of the field, trace the evolution of knowledge, and identify influential journals, seminal works, leading institutions, and contributing countries. Second, few studies have combined bibliometric mapping with qualitative synthesis to bridge technical, psychological, and educational perspectives on the university student sample. Finally, although AI-based chatbots offer accessible and scalable mental health support for university students, important long-term considerations, such as ethical challenges (Chan 2025 ), and potential development in this field should be taken into consideration. As these tools become increasingly integrated into student life, moving beyond short-term evaluations and exploring future directions with a focus on sustainability, equity, and long-term well-being outcomes (Nations 2015 ) is essential. There is a need for systematic bibliometric analysis to uncover the structure of research output and for qualitative synthesis to identify overarching gaps and future research pathways in this interdisciplinary domain. 3. Method This study employs a dual-method approach integrating bibliometric analysis and systematic literature review (SLR). Bibliometric analysis systematically maps knowledge domains and citation networks, revealing publication/citation trends, key topics, thematic evolution, and influential sources, countries, institutions, and documents (Donthu et al. 2021 ; Z. Liu et al. 2015 ). SLR identifies future directions and provides qualitative depth. These methods are mutually reinforcing, and their integration is increasingly advocated for robust research synthesis (Marzi et al. 2025 ), as demonstrated in prior studies (Abuhassna et al. 2024 ). 3.1 Search strategies We searched the Web of Science (WoS) Core Collection and Scopus databases for peer-reviewed English-language articles up to 5 July, 2025. Search terms combined AI/chatbot-related keywords (e.g., “AI chatbot,” “conversational agent,” “GPT-4”), mental health terms (e.g., “mental health,” “well-being,” “depression,” “anxiety,” “stress”), and student context terms (e.g., “university student,” “college student,” “higher education,” “undergraduate”). Boolean operators (AND/OR) and wildcards were used to capture variations. The initial search yielded 417 records (148 from WoS, 269 from Scopus). We excluded duplicates, non-article document types (reviews, conference papers, etc.), and non-English works. We then screened titles, abstracts, and keywords for relevance to the intersection of AI/chatbots, mental health, and university students. This process (see PRISMA flow diagram in Fig. 1 ) resulted in 58 articles for bibliometric analysis. From these, we identified the 20 most-cited empirical papers (excluding reviews) as the basis for our systematic literature review. 3.2 Data analysis For bibliometric analysis, we used the R bibliometric package, Biblioshiny and VOSviewer software. We examined annual publication counts, author and country collaborations, and distribution of publications by source, country, institution, and document. We identified the most cited documents and conducted co-citation analysis to map the intellectual base. Keyword co-occurrence analysis and trend-topic analysis were performed to detect major themes and their evolution. For the systematic review, we categorized the top 20 cited papers into thematic clusters (derived from keyword co-occurrence). For each paper, we extracted information on research design, objectives, key findings, limitations, and proposed future directions. Synthesizing these data across clusters, we identified the recommended future research directions. 4. Results 4.1 Global publication trends We identified 58 relevant articles published between 2012 and July 2025, with an average document age of 1.74 years. Annual output remained very low until 2021 but then grew rapidly (Fig. 2 ). The first relevant publication appeared in 2012, followed by a hiatus until 2017. From 2018–2021, only 1–5 papers appeared annually. Beginning in 2022, publication counts increased sharply: 4 papers in 2022, 8 in 2023, 12 in 2024, and 25 by July 2025. Overall growth accelerated after 2021, reflecting heightened research interest. This trend suggests that AI-based student mental health chatbots are an emerging topic, which may be due to the improvements in AI technology and increased student mental health concerns (e.g., due to the COVID-19 pandemic). 4.2 Analysis of the most influential and productive agents (countries, affiliations, sources and documents) Countries and Institutions: Thirty-three countries contributed to the 58 articles. The result was based on co-authorship data, as publications often involve international collaborations, the sum of country contributions exceeds the total number of articles China was listed as a contributing country in 46 of the 58 publications, followed by South Korea (19), the USA (19), Canada (14), Australia (12), Jordan (12), and the UK (11) (Table 1 ). Notably, countries with strong publication counts include both traditional research powers and emerging academic communities (e.g., South Africa, Saudi Arabia, Lebanon with 6–10 each). International collaboration networks (Fig. 3 ) show strong links between the U.S. and Canada and within the Middle East, as well as connections among China, Australia, and the UK. Nevertheless, the overall international co-authorship rate was only 25.4%, indicating space for broader collaboration. Table 1 Top 10 countries and institutions for AI–based chatbots in mental support among university students. Rank Country Counts Affiliation Counts 1 CHINA 46 The University of British Columbia (Canada) 10 2 SOUTH KOREA 19 Macquarie University (Australia) 8 3 USA 19 The University of Jordan (Jordan) 7 4 CANADA 14 Ming Chuan University (Taiwan,China) 5 5 AUSTRALIA 12 University of Huddersfield (UK) 5 6 JORDAN 12 Hanyang University (South Korea) 4 7 UK 11 Hubei University of Chinese Medicine (China) 4 8 SOUTH AFRICA 10 Imperial College London (UK) 4 9 SAUDI ARABIA 7 National Tsing Hua University (Taiwan,China) 4 10 LEBANON 6 Seoul National University (South Korea) 4 The University of British Columbia (Canada) was the most productive institution (10 papers), followed by Macquarie University (Australia, 8) and the University of Jordan (7). Other notable institutions (4–5 papers each) were primarily in East Asia (e.g., Ming Chuan University, Hanyang University, Hubei University of Chinese Medicine, National Tsing Hua University, Seoul National University) and the UK (University of Huddersfield, Imperial College London). The top institutions span North America, Asia, Australia, the UK, and the Middle East, reflecting growing global interest, more details can be found in Table 1 . Journals: The 58 papers appeared in 46 different journals. The Journal of Medical Internet Research (JMIR) published the most (4 papers), followed by Digital Health and JMIR Mental Health (3 each). Among top journals, JMIR MHealth and UHealth had the highest impact factor (6.1), followed by JMIR (6.0) and JMIR Mental Health (5.8). Most top outlets were in Q1, indicating high source quality (Table 2 ). Table 2 Top 10 journals in publications on AI–based chatbots related to mental support among university students. Rank Journal Count IF Quartile 1 JOURNAL OF MEDICAL INTERNET RESEARCH 4 6 1 2 DIGITAL HEALTH 3 3.3 1 3 JMIR MENTAL HEALTH 3 5.8 1 4 INTERNATIONAL JOURNAL OF HUMAN-COMPUTER INTERACTION 2 4.9 1 5 INTERNET INTERVENTIONS-THE APPLICATION OF INFORMATION TECHNOLOGY IN MENTAL AND BEHAVIOURAL HEALTH 2 4.1 1 6 JMIR FORMATIVE RESEARCH 2 2.1 3 7 JMIR MHEALTH AND UHEALTH 2 6.1 1 8 JOURNAL OF AFFECTIVE DISORDERS 2 4.9 1 9 ACM TRANSACTIONS ON APPLIED PERCEPTION 1 2.1 3 10 APPLIED SCIENCES-BASEL 1 2.5 3 Key Documents: The most-cited article was “Delivering Cognitive Behavior Therapy to Young Adults With Symptoms of Depression and Anxiety Using a Fully Automated Conversational Agent (Woebot): A Randomized Controlled Trial” by Fitzpatrick et al. (2017b) with 1,012 citations. This landmark Woebot RCT demonstrated feasibility and preliminary efficacy of an automated CBT chatbot for college students. It was followed by Fulmer et al. ( 2018 ) ( Using Psychological AI (Tess) , 284 cites), which showed feasibility of an AI chatbot to reduce depression/anxiety symptoms, and Park et al. ( 2019 ) ( Designing a Chatbot...Brief Motivational Interview ), with 99 cites (Table 3 ). These top studies established foundational evidence in the field. We also analyzed the co-citation network, Fitzpatrick et al. ( 2017a ), Kroenke et al. ( 2001 ), Vaidyam et al. ( 2019 ) and Auerbach et al. (2018). These articles in the co-citation network (Fig. 4 ) laid the foundation for the development of the research field on AI-based chatbots in support of mental health among university students. Table 3 Top 10 most cited documents Paper DOI Total Citations TC per Year Normalized TC (Fitzpatrick et al. 2017b) 10.2196/mental.7785 1012 112.44 1.89 (Fulmer et al. 2018 ) 10.2196/mental.9782 284 35.50 1.00 (Park et al. 2019 ) 10.2196/12231 99 14.14 1.00 (H. Liu et al. 2022 ) 10.1016/j.invent.2022.100495 95 23.75 2.77 (Gabrielli et al. 2021 ) 10.2196/27965 81 16.20 1.56 (Abdaljaleel et al. 2024 ) 10.1038/s41598-024-52549-8 70 35.00 5.45 (Zhou et al. 2021 ) 10.1177/1357633X211047285 66 13.20 1.27 (Klos et al. 2021 ) 10.2196/20678 65 13.00 1.25 (Sebastian and Richards 2017 ) 10.1016/j.chb.2017.03.071 58 6.44 0.11 (Potts et al. 2021b) 10.1007/s41347-021-00222-6 31 6.2 0.6 4.3 Key words and trend topics The most frequent keywords included “ChatGPT,” “AI,” “student anxiety,” “depression,” “stress,” and “mental health”, as revealed by the word cloud visualization (Fig. 5 ). Our trend topic analysis (Fig. 6 ) shows that “conversational agents” has been a consistent focus throughout the period. Notably, terms like “AI” and “ChatGPT” became prominent around 2023, reflecting the impact of generative AI breakthroughs (e.g., ChatGPT release in late 2022). The emergence of “mHealth” as a trending topic in 2023 indicates growing integration with mobile technologies. Terms “female” and “student” appearing since 2021 suggest increasing attention to gender-specific aspects. “Stress” began trending in 2022, highlighting heightened interest in using chatbots for stress reduction. Co-occurrence analysis (Fig. 7 ) identified 56 items meeting the threshold of at least 2 co-occurrences. The network visualization reveals keyword clusters, each representing distinct thematic groups, where larger font sizes denote greater prominence. Of these, 57 items were grouped into 5 clusters. Across all clusters, keywords such as chatbot , mental health , academic stress , anxiety , AI , depression, and university student emerged as highly salient. The interconnections within each cluster further delineate subthemes under the broader research focus. We identified the top plus key words under each cluster and considered the high-frequency words and the content of the individual articles to find potential subtopics for further systematic review analysis. The contents of the five clusters are presented in Table 4 . Cluster 1 emphasized the AI chatbot for university students’ academic stress; cluster 2 was labeled for chatbot prevention or treatment, such as depression and anxiety symptoms, in the high-risk group; cluster 3 highlighted chatbots for self-help with emotional or mental health issues; cluster 4 focused on evaluations of chatbot intervention; and cluster 5 focused on the user experience or perceptions of chatbots. Table 4 Sub-themes clusters Cluster # Top keyword plus Potential topic 1 Academic stress, achievement, stress management, performance Chatbot in stress setting 2 Mental health, mental disorder, illness, symptoms, anxiety, depression Chatbot for high-risk group 3 phq-9, loneliness, self-disclosure, emotion, self-help, well-being, Chatbot for emotion support 4 randomized controlled trial, questionnaire, validation, scale, intervention, effectiveness Research design and intervention evaluation 5 User experience, conversational agent, human computer interaction, use intension Chatbot technology and user experience 4.4 Future directions based on a systematic literature review After confirming the subthemes, we conducted a systematic review and investigated the top 4 cited articles under each cluster in terms of their method, objective and future directions. The outlines of these 20 articles are presented in Table 5 . Future directions are highlighted below. Table 5 Outline of the systematic literature review Cluster Article Method Objective(s) Future Direction 1 (Nelekar et al. 2022 ) Nelekar et al., 2022 Quasi experimental Evaluate the ECA’s effectiveness in reducing academic stress among Indian students. 1. Longitudinal Studies 2. Larger/Diverse Samples 3. Enhanced Personalization 4. Expanded Content 5. Integration with Institutions 6. Cross-Cultural Validation 7. Behavior Tracking 1 (Kavakli et al. 2012 ) Kavakli et al., 2012 Qualitative and exploratory Explore how interaction design can be used to create an artificial therapist capable of supporting mental health interventions, with a particular focus on stress management for university students. 1. Develop higher-fidelity prototypes 2. Conduct empirical studies with real users and mental health practitioners 3. Incorporate emotion detection and adaptive feedback 4. Explore long-term user engagement 5. Investigate multimodal interfaces 1 (Moldt et al. 2022 ) Moldt et al., 2022 mixed-methods research design Evaluate the effectiveness of a chatbot in assessing the stress levels of medical students(N = 284) in everyday conversations and identify the main condition for accepting a chatbot as a conversational partner based on validated stress instruments 1. Improve Chatbot Technical Performance 2. Develop a User-Centered Bot Design 3. Increase Transparency in Bot Functionality 4. Enhance Psychometric Validity of Measurement 5. Expand Cultural and Linguistic Scope 6. Contribute to AI and Education Research Communities 1 (Maciejewski and Smoktunowicz 2023 ) Maciejewski & Smoktunowicz, 2023 parallel randomized controlled trial (RCT) with two conditions compared on three time points Determine whether a psychological self-guided intervention could be effective when delivered with a low-effort approach through a chatbot on Meta's Messenger among 372 Polish university students. 1. Include Active Control Comparisons 2. Enhance Measurement of Engagement 3. Adopt Participatory Design Approaches 4. Explore Multimodal Interaction Features 5. Evaluate AI-Enhanced Chatbots 6. Optimize Intervention Intensity 7. Incorporate Booster Sessions or Follow-ups 2 (Fitzpatrick et al. 2017b) Fitzpatrick et al., 2017 randomized controlled trial (RCT) design The study's goal was to assess the feasibility, acceptability, and preliminary efficacy of a completely automated conversational agent for delivering a self-help program to college students who self-identify as suffering from anxiety and depression. 1. Enhanced Trial Design 2. Engagement Tracking 3. Population Expansion 2 (Fulmer et al. 2018 ) Fulmer et al., 2018 randomized controlled trial experimental study Assess the feasibility and usefulness of utilizing Tess, an integrative psychological AI, to reduce self-identified symptoms of depression and anxiety in college students(n = 75). 1. Test in clinical population 2.Compare the control group against Traditional therapy, Teletherapy, VR/interactive courses 3.Integrate multimodal emotion recognition 4.Ethical safeguard 2 (Klos et al. 2021 ) Klos et al., 2021 pilot randomized controlled trial assess the viability, acceptability, and possible impact of utilizing Tess, a chatbot, for assessing symptoms of depression and anxiety in university students. 1. larger samples 2. Waitlist controls 3.Stratified/unbalanced randomization 4. Participant diversity 5. Intervention design (Therapy-specific chatbots, Hybrid models) 6. Comparative research 2 (Y. Li et al. 2025 ) Li et al., 2025 single-group pretest-posttest study evaluate the feasibility, acceptability, safety, and preliminary efficacy of a chatbot-based MBSR intervention for university students with depressive symptoms(N = 30) larger, controlled sample, extending the follow-up period, and collecting data on confounding factors to better assess the lasting effects of the intervention 3 (H. Liu et al. 2022 ) Liu et al., 2022 randomized controlled trial This study compared chatbot therapy to well-established bibliotherapy to gather evidence for chatbot therapy's effectiveness as a convenient, low-cost, interactive self-help intervention for depression (n = 52). 1. Technical enhancements 2. Content expansion 3. Diverse demographics 4. Conduct longitudinal studies 3 (Salamanca-Sanabria et al. 2023 ) Salamanca-Sanabria et al., 2023 Interview examine how university students(N = 30) and mental health advocates in Singapore perceive mental health services, awareness campaigns, and mHealth interventions, with particular emphasis on the acceptability of conversational agent technologies for the prevention of anxiety and depression 1. Targeting holistic wellbeing 2. Adapting content to local and contextual barriers 3. Collaborating with government and local organizations 4. Carefully managing the use of incentives 5. Expanding research across different populations and cultural contexts 3 (Hopman et al. 2023 ) Hopman et al., 2023 Quasi experimental/Pilot study/Pre-post design without control group Examines whether an ECA can successfully deliver a highly targeted cognitive emotion regulation psychoeducation intervention(N = 138) 1. Implement Controlled Experimental Designs 2. Evaluate Long-Term Impact 3. Test with the Target Clinical Population 4. Enhance inclusivity through Personalization 5. Gather Feedback from High-Rating Users 6. Refine Conversation Design 7. Adapt ERICA for MHealth Delivery 8. Engage Stakeholders in Codesign 3 (Potts et al. 2021a ) Potts et al., 2021b Qualitative research using Living Labs methodology This study aims to codesign a chatbot that supports mental wellbeing among people in rural areas (include university students’ sample) 1. Advance chatbot capabilities 2. Integrate machine learning techniques 3. Address the challenge of empathetic dialog 4. Develop multidisciplinary frameworks 5. Reimagine chatbot roles as early detection and self-monitoring tools 6. Explore personalization and ethical design 4 (Gabrielli et al. 2021 ) Gabrielli et al., 2021 Mixed Methods Proof-of-Concept Study conduct a proof-of-concept evaluation measuring the engagement and effectiveness of Atena, a psychoeducational chatbot supporting healthy coping with stress and anxiety, among a population of university students(N = 71) 1. Rigorous Efficacy Testing 2. Enhanced Engagement Measurement 3. User-Centered Development: Employ user-centered methodology 4. Leverage Contextual Insights 4 (Sebastian and Richards 2017 ) Sebastian & Richards, 2017 Quasi-experimental pretest-posttest design Examine whether education or contact interventions delivered by either ECAs or video presentations result in improvements in Mental Health Literacy postintervention compared to preintervention among the university students(N = 245). 1. Longitudinal Research 2. Clarify Stigmatization Process 3. Expand Intervention Scope 4. Enhance Agent Personalization 4 (Vereschagin et al. 2024 ) Vereschagin et al., 2024 A 2-arm, parallel-assignment, single-blinded, 30-day randomized controlled experimental research design examine the effectiveness of the Minder mobile app in improving mental health and substance use outcomes in a general population of university students (N = 1489) 1. Conduct Per-Protocol and Subgroup Analyses 2.Personalization of Content and Features 3. Enhance User Engagement 4. Implement Co-development and Participatory Design 5. System Integration 4 (Trappey et al. 2022 ) Trappey et al., 2022 Module development and quasi experimental pretest-posttest design Development of an Empathy-Centric Counseling Chatbot System and Evaluate empathy-centric counseling in the chatbot system among the university student in Taiwan 1.Enhance the therapeutic realism of the system 2.Integrate more advanced language modeling 3.Expand intervention modules, and explore VRECC’s potential as a scalable virtual assistant for campus-based mental health services 5 (Park et al. 2019 ) Park et al., 2019 Qualitative Case Study Create a conversational sequence for a quick motivational interview to be delivered by a Web-based text messaging application (chatbot) and study its conversational experience with graduate students in their coping with stress. 1. Explore MI Sequences 2. Multi-Session Interventions 3. Larger Field Studies 4. Develop Assessment Toolkit 5. Incorporate Multimedia/Embodiment 5 (Ehrlich et al. 2024 ) Ehrlich et al., 2024 A two arm, randomized, controlled trial Evaluate the Mind Tutor, an AI-enhanced tool created to boost student well-being. 1. Link Usage Patterns to Outcomes 2. Understand User Interaction 3. Tailor App Development 5 (Gbollie et al. 2023 ) Gbollie et al., 2023 Exploratory factor analysis and multivariate ordinal regression examine university students ( n = 17 838)’ experiences with, attitudes toward, and intentions to use digital mental health solutions, and to identify key factors influencing their intention to use these tools. 1. Investigate low engagement with chatbots and develop strategies to improve uptake 2. Develop psychoeducational interventions 3. Expand research samples and national representation 5 (Abdaljaleel et al. 2024 ) Abdaljaleel et al., 2024 structured survey Compare college students' opinions of counseling offered by human counselors to counseling generated by Pi, a helpful and sympathetic chatbot. 1. Identify Key Factors led to the positive perception and inability to distinguish AI from human 2. Complex & Extended Interactions 3. Investigate AI's Multimodal Communication (facial expressions, vocal tones) Cluster 1: Chatbot use for academic stress management is still limited by technology. Future research must focus on advancing NLP, emotion recognition, and conversational flow. Personalizing interventions to individual user needs is also critical. Studies require larger, more diverse samples and longitudinal designs to assess sustained impact. Collaborating with universities to integrate chatbots into student support systems is recommended. Cluster 2: Future research should utilize clinically confirmed samples to address self-selection bias and clarify therapeutic boundaries for diagnosed populations. Direct comparisons with traditional therapy are essential to assess efficacy. Advancing multimodal interactions (text, voice, visuals) remain crucial for engagement. Strict adherence to ethical standards, particularly regarding privacy, informed consent, and safeguarding is imperative. Cluster 3: Future research must prioritize participatory design, actively involving both students and mental health professionals. Given demonstrated cross-cultural variations in effectiveness, cultural and contextual adaptation is essential. Ensuring acceptability and efficacy requires ongoing user feedback and broad-scale testing. Developing scalable, culturally sensitive systems is key for wider implementation. Cluster 4: Rigorous experimental designs including control groups, blinding, stratified randomization, and longitudinal follow-ups are fundamental for robust evaluation. Interventions should adopt multi-session, long-term approaches to measure sustained outcomes. Improved metrics for engagement, fidelity, and satisfaction are needed, alongside mixed methods approach. Integrating novel technologies (e.g., VR, motivational interviewing, hybrid human-AI) can enhance interventions. Developing modular systems targeting specific psychological goals (e.g., emotion regulation) is recommended. Cluster 5: Future research must incorporate sociocultural factors in design and evaluation, as user experience varies significantly across cultures. Despite this variation, enhancing engagement through technological innovation is consistently important. Understanding determinants of positive experience and mitigating factors for low retention is critical. Chatbots supporting sustained, context-aware interactions via multimodal communication, personalization, and privacy features enhance satisfaction and long-term adherence. 5. Discussion 5.1 Overall findings from bibliometric analysis The evolution of publication volume over time closely reflects the technological advancements that underpin the development of AI-based chatbots. One of the earliest relevant studies, conducted by Kavakli et al. ( 2012 ), emphasized the potential of human‒computer interaction (HCI) to enable “virtual humans” to act as virtual psychologists. However, at that time, the field of AI had not yet focused on modeling expressive agent motions capable of conveying personality and nonverbal communication effectively. A technological shift occurred approximately 2017–2018, marked by improvements in voice recognition and conversational interfaces, which made conversational agents more accessible and functional (Fitzpatrick et al. 2017b). This period also resulted in a corresponding increase in related publications. A major turning point arrived at the end of 2022 with the rapid development of generative large language models (LLMs) (Khan et al. 2024 ), which significantly increased the capacity for personalized interaction. Many of the reviewed studies acknowledged this technological leap, engaged in natural, multiturn conversations that emulated human empathy, understanding, and responsiveness (Al-Amin et al. 2024 ). These advances, combined with the mental health challenges exacerbated by the COVID-19 pandemic (Aristovnik et al. 2020 ), may contributed to a sharp rise in publications after 2021, consistent with global reports of increased interest in digital mental health solutions during and after the pandemic (He et al. 2022 ; Marwaha et al. 2025 ). In terms of research geography, China, South Korea, and the United States lead in total publication volume. However, no institutions from these countries rank among the top three most productive institutions, suggesting relatively limited domestic and inter-institutional collaboration. In contrast, institutions from Canada, Australia, and Jordan are among the most productive, indicating robust national-level and interinstitutional collaboration. Current research remains concentrated in developed and upper-middle-income countries, with minimal representation from low-income or underdeveloped regions. This geographic imbalance mirrors the need for more inclusive and geographically diverse research to address disparities in access to technological health interventions (Hoagland and Kipping 2024 ). The prominence of “AI” and “ChatGPT” as the top keywords confirms that generative AI, particularly ChatGPT has become the most frequently studied chatbot in this field. Alongside these developments, mobile health (mHealth) applications have also attracted notable attention, indicating a broader trend toward technological integration in mental health care. High-frequency mental health-related keywords such as “anxiety” and “depression” align with findings from recent studies during or shortly after the COVID-19 pandemic, when the prevalence of these conditions reportedly rose by about 25% worldwide (WHO, 2022). The pandemic heightened concerns on student well-being and normalized mental health discourse within higher education (Harris et al. 2022 ). The recurring term “stress management” underscores the widespread presence of academic stress among students and the recognition of chatbots as potential tools for its alleviation (Mofatteh, 2021 ). Collectively, these keyword patterns emphasize the centrality of stress, anxiety, and depression in both developed and developing contexts, and highlight the importance of embedding digital technologies into campus-based prevention and intervention strategies (S. Zhai et al. 2025 ). The frequent appearance of “intervention” further suggests that much of the literature is experimental or quasi-experimental in nature, reflecting the rapidly evolving state of AI-based mental health applications in higher education. 5.2 Future directions related to SDGs based on the systematic literature review and beyond Based on systematic literature review and the preceding bibliometric analysis, the following future research directions integrate micro-level recommendations, focused on research design, technology, user engagement, and cultural adaptation, with macro-level strategies, such as global collaboration, institutional integration, and ethical governance. Framing these directions within the United Nations SDGs highlights their relevance not only to academic and technological advancement but also to sustainable global well-being. From the perspectives of research design and intervention evaluation, many studies emphasize the need for more rigorous experimental designs. Research design shift strengthens the evidence base for effective mental health interventions in higher education, which aligns with SDG 3 (Good Health and Well-being) and SDG 4 (Quality Education). Randomized controlled trials (RCTs) are widely regarded as the most reliable methods for establishing causal relationships (Krauss 2021 ). Meanwhile, we should conduct more longitudinal studies to track long-term effects and psychological and educational research, where short-term or single-session interventions without follow-up often limit the generalizability and durability of the results. Furthermore, although some studies have incorporated user feedback through qualitative methods or postintervention surveys, the number of mixed-method studies remains limited. Expanding the use of mixed-method designs could enrich the evaluation of both the psychological impact and the user experience. From a technological perspective, there is a need for technology improvement and make breakthroughs in generative AI models. This direction supports SDG 9 (Industry, Innovation, and Infrastructure) and SDG 3, fostering innovation that directly benefits mental health outcomes. As chatbots become more advanced, future studies should focus on enhancing natural language processing (NLP), emotion recognition, and multimodal interaction capabilities to make interventions more responsive, empathetic, and context aware (Babu et al. 2024 ). With respect to design-related concerns, a recurring issue is low user engagement, particularly in mental health apps, such as that mentioned by Ehrlich et al. ( 2024 ), poor engagement not only limits the effectiveness of interventions but also hampers efforts to evaluate outcomes meaningfully. A frequently reported form of user complaint is the lack of personalization, as Fitzpatrick et al. (2017b) attempted to explore how to improve user engagement via personalization, and the latest studies recommended analyzing user input and behavioral data (Manole et al. 2024 ) or tailor interaction (Ehrlich et al. 2024 ). Some studies have proposed participatory or codesign approaches as potential solutions, allowing users to contribute to the design and refinement of chatbot functions, content, and delivery style (Grové 2021 ; Potts et al. 2021b). Additionally, the review indicates that future development of AI-based chatbots should consider the local sociocultural background of target populations, not only by translating chatbot content into local languages but also by aligning tone, values, and support strategies with cultural expectations and norms (C. Zhai and Wibowo 2022 ). These strategies offer valuable insights into the future development of more adaptive and engaging mental health applications, these strategies advance SDG 10 (Reduced Inequalities), ensuring equitable and culturally relevant support. From a macro perspective, it is essential to reduce the global divide in AI research and implementation. The World Health Organization (2020) has highlighted the potential of AI to bridge health gaps in low- and middle-income countries (LMICs). The current Bibliometric evidence indicates that research output is concentrated in a limited number of technologically advanced countries, creating a risk of widening global disparities in access to AI-based mental health tools. Such issues directly address SDG 17 (Partnerships for the Goals) and SDG 10 (Reduced Inequalities), promoting inclusive and cooperative innovation (Nations 2015 ). Alami et al. ( 2020 ) further proposed training and retention of local expertise, a comprehensive monitoring system, a systems-based approach to implementation, and accountable local leadership inclusive of all stakeholders. Therefore, international partnerships through joint projects, shared datasets, and capacity-building programs is need (Mhlanga 2023 ). Beyond geographical inequities, health equity must also be extended to marginalized university student groups such as ethnic minorities, immigrants, people with disabilities, and sexual minorities. For example, Nadarzynski et al. ( 2024 ) noted that stigma and discomfort often hinder these populations from discussing sensitive issues, highlighting the importance of culturally sensitive chatbot design. Another critical future direction is strengthening the cooperation network between AI chatbot developers and universities. Many reviewed studies have emphasized the need for campus-specific applications that address the daily experiences and psychosocial challenges of students (Ehrlich et al. 2024 ; Vereschagin et al. 2024 ). For some mental health App developers should work closely with university counseling centers to facilitate ethically responsible data collection, crisis sample screening, and timely interventions for students seeking psychological support (Lattie et al. 2022 ). In parallel, collaboration with university administrations is essential to promote awareness campaigns, ensure institutional endorsement, and foster the adoption of AI-based tools among students. Moreover, chatbot-related competencies could be integrated into mental health programs and curricula, equipping students with skills to effectively use digital tools for self-help and preventive care, students who have received sufficient mental health education are better prepared to make educated healthcare decisions in the future, stressing the long-term influence of educational interventions (S. Zhai et al. 2025 ). Finally, developers should leverage the research resources available within universities, including longitudinal student data and campus-based pilot studies, to enable evidence-driven refinement and monitoring of interventions. Such systemic integration would not only enhance the ecological validity of chatbot applications but also advance SDG 3 (Good Health and Well-being) by improving access to mental health care, and SDG 4 (Quality Education) by embedding innovative support systems into higher education environments. Ethical issues should be greatly concerned regarding deployment of AI in mental health. Many reviewed studies mentioned ethical safeguard, privacy, data security, and algorithmic bias are particularly pressing in the context of vulnerable populations such as university students (Fulmer et al. 2018 ; Olla et al. 2025 ). To deal with this issue, it’s recommended to obey data privacy laws where necessary, or make sure that user know how their data is being used, whether that means just storing it or using it to train AI models, or making sure that patients know how their data is being stored and kept safe (Coghlan et al. 2023 ). Moreover, the cases are more complex than we expected, some contended that users could be discouraged from obtaining appropriate mental health services if they receive inappropriate responses (Vilaza and McCashin 2021 ), or users’ dependence and liable to manipulation on the chatbots (Laestadius et al. 2022 ). Coghlan et al. ( 2023 ) mentioned five AI ethics principles, (a) non-maleficence, (b) beneficence, (c) respect for autonomy, (d) justice and (e) explicability. These principles are quite relevant to the AI technology design and the protection from the law, which also need the supervision from the human (Coghlan et al. 2023 ), however, this topic is still underdiscussed and need to be further researched. Although some governments established some policies to monitor (Martinez-Martin 2023 ), but AI Chabot still has a lot of room for improvement on some topics such as the boundaries of psychological interventions, the discussion of sensitive topics, and the standardization and rationalization of the accountability system (Vilaza and McCashin 2021 ), and it is also necessary for policy makers and developers, psychologists, legal practitioners, university administrators, and university users to participate in the supplementation of the ethical framework of the AI-chatbot together. These priorities align with SDG 16 (Peace, Justice, and Strong Institutions), which underscores the importance of trustworthy governance, and with SDG 3, as ethical safeguards are integral to protecting student well-being. Building trust through ethical governance will not only enhance user acceptance but also ensure that AI chatbots contribute positively to mental health support in ways that are safe, equitable, and sustainable. 5.3 Implications This study advances theoretical understanding of digital mental health and AI-based chatbots within higher education context. By integrating bibliometric mapping with a systematic literature review, it bridges technological perspectives from computer science with psychological and educational frameworks. This study reveals emerging affective dimensions in human‒chatbot interactions, suggesting an expansion of traditional human‒computer interaction frameworks to incorporate digital empathy, trust-building, and emotional regulation. Importantly, the study situates these contributions within the Sustainable Development Goals (SDGs), particularly SDG 3 (Good Health and Well-being) and SDG 4 (Quality Education), SDG 10 (Reduced Inequalities), SDG 16 (Peace, Justice, and Strong Institutions) and SDG 17 (Partnerships for the Goals) reinforcing the role of AI-based mental health tools in supporting equitable access to psychological support for university students. The findings suggest that policymakers should consider integrating chatbots as a complement to traditional services into institutional mental health strategies (Olla et al., 2025 ). They should actively encourage the use of therapeutic chatbots as a complement to traditional campus services, providing professional guidance from mental health practitioners to help students identify high-quality resources and utilize them effectively for psychological well-being (Gao 2025 ; Oghenekaro and Okoro 2024 ). Also, they should establish regulatory frameworks that ensure privacy, data protection, and ethical accountability in AI-driven mental health tools. Promoting equitable funding mechanisms and capacity-building programs. Researchers should establish a research collaboration network and design rigorous experimental studies including randomized controlled trials (RCTs), longitudinal research, cross-disciplinary innovation and cultural adaptation research. For developers and technology providers, they should enhance NLP capabilities, multimodal interaction, and adaptive learning systems, engage in participatory co-design with students, educators, psychologists, counselors and lawyers. International organizations should facilitate global partnerships and knowledge exchange, supporting LMICs in adopting AI for mental health, they should also set ethical benchmarks and foster open-access research databases, ensuring global inclusivity. 5.4 Limitations This study is based on data retrieved from the Web of Science Core Collection and Scopus, which ensures a focus on high-quality publications. However, this also means that relevant studies indexed in other databases, such as PubMed, may have been excluded. As a result, the overall number of analyzed publications may be limited, potentially affecting the comprehensiveness of the findings. Furthermore, our systematic review focused on highly cited papers from bibliometric analysis. This strategy may have excluded newer or lower-cited studies, meaning some emerging insights could be missed. Consequently, the future directions identified are drawn from influential works and should be interpreted in light of this selection bias. Additionally, the exclusion of non-English language articles means that studies published in other languages, such as Chinese or Spanish, were not included. This could lead to a regional bias in the interpretation of research distribution, particularly underrepresentation of contributions from countries such as China and those in Latin America. Finally, the current study focused only on university students, which limits its generalizability to other populations, such as clinical samples or broadened age groups. 6. Conclusion This study combined bibliometric analysis and a systematic literature review to explore the landscape of AI-based chatbot research in supporting university students' mental health. The findings revealed a growing interest in this field, with emerging research clusters focusing on academic stress, emotional support, high-risk mental health scenarios, intervention design, and the user experience. Our findings highlight the link between digital mental health interventions and SDGs and practical implications for higher education. Researchers should pursue more longitudinal and cross-cultural studies, policymakers and administrators need to establish supportive regulations and institutional frameworks. Developers must ensure inclusive design and data transparency. Moving forward, global collaboration is indispensable. This emerging research area associates with key Sustainable Development Goals, and continued development has the potential to contribute to student well-being and broader SDG targets. Declarations Funding No funding applicable Data Availability All data generated or analyzed during this study are included in this published article and its supplementary materials. Raw bibliometric data were retrieved from the Web of Science and Scopus database. Additional data are available from the corresponding author upon reasonable request. Competing Interests The authors declare that they have no competing interests. Ethical Statements This article does not contain any studies with human participants or animals performed by any of the authors. Ethical approval was therefore not required. Author Contribution All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Yi Xing, Chen Ying, Loh Sau Cheong and Siaw Yan Li. The first draft of the manuscript was written by Yi Xing and all authors commented on previous versions of the manuscript. The final version was proofread by Hao Li Jie and Hassan Abuhassna. All authors read and approved of the final manuscript. Data Availability All data generated or analyzed during this study are included in this published article and its supplementary materials. Raw bibliometric data were retrieved from the Web of Science and Scopus database. Additional data are available from the corresponding author upon reasonable request. References Abdaljaleel M et al (2024) 'A multinational study on the factors influencing university students’ attitudes and usage of ChatGPT', Scientific Reports , 14 (1), 1983. https://doi.org/10.1038/s41598-024-52549-8 Abuhassna H et al (2024) The Information Age for Education via Artificial Intelligence and Machine Learning: A Bibliometric and Systematic Literature Analysis. Int J Inform Educ Technol 14:700–711. https://doi.org/10.18178/ijiet.2024.14.5.2095 Adamopoulou E, Moussiades L (2020) An Overview of Chatbot Technology. In: Maglogiannis I, Iliadis L, Pimenidis E (eds) Artificial Intelligence Applications and Innovations, 2 edn. Springer International Publishing), Greece, pp 373–383. https://doi.org/10.1007/978-3-030-49186-4_31 Neos Marmaras Al-Amin M et al (2024) 'History of generative Artificial Intelligence (AI) chatbots: past, present, and future development', arXiv preprint arXiv:2402.05122 Alami H et al (2020) Artificial intelligence in health care: laying the Foundation for Responsible, sustainable, and inclusive innovation in low- and middle-income countries. Globalization Health 16(1):52. https://doi.org/10.1186/s12992-020-00584-1 Aristovnik A et al (2020) Impacts of the COVID-19 pandemic on life of higher education students: A global perspective. Sustainability 12(20):8438. https://doi.org/10.3390/su12208438 Auerbach, Randy P et al (2018) WHO world mental health surveys international college student project: Prevalence and distribution of mental disorders. J Abnorm Psychol 127(7):623. https://doi.org/10.1037/abn0000362 Babu A et al (2024) 'Multimodal Emotion Analysis Using Integrating NLP, AI, and Facial Expression Recognition for Enhanced Emotion Detection', 2024 IEEE International Conference on Signal Processing, Informatics, Communication and Energy Systems (SPICES) , 1–6. https://doi.org/10.1109/SPICES62143.2024.10779754 Bansal G et al (2024) Transforming conversations with AI—A comprehensive study of ChatGPT. Cogn Comput 16(5):2487–2510. https://doi/10.1007/s12559-023-10236-2 Chan CKY (2025) AI as the Therapist: Student Insights on the Challenges of Using Generative AI for School Mental Health Frameworks. Behav Sci (Basel) 15(3). https://doi.org/10.3390/bs15030287 Coghlan S et al (2023) To chat or bot to chat: Ethical issues with using chatbots in mental health. Digit Health 9:20552076231183542. https://doi.org/10.1177/20552076231183542 Denecke K, Gabarron E (2024) The ethical aspects of integrating sentiment and emotion analysis in chatbots for depression intervention. Front Psychiatry 15:1462083. https://doi.org/10.3389/fpsyt.2024.1462083 Donthu N et al (2021) How to conduct a bibliometric analysis: An overview and guidelines. J Bus Res 133:285–296. https://doi.org/10.1016/j.jbusres.2021.04.070 Ehrlich C et al (2024) Evaluation of an Artificial Intelligence Enhanced Application for Student Wellbeing: Pilot Randomised Trial of the Mind Tutor. Int J Appl Posit Psychol 9(1):435–454. https://doi.org/10.1007/s41042-023-00133-2 Eltahawy L et al (2024) Can robots do therapy? Examining the efficacy of a CBT bot in comparison with other behavioral intervention technologies in alleviating mental health symptoms. Computers Hum Behavior: Artif Hum 2(1):100035. https://doi.org/10.1016/j.chbah.2023.100035 Farzan M et al (2025) Artificial intelligence-powered cognitive behavioral therapy chatbots, a systematic review. Iran J psychiatry 20(1):102. https://doi.org/10.18502/ijps.v20i1.17395 Feng Y et al (2025) Effectiveness of AI-Driven Conversational Agents in Improving Mental Health Among Young People: Systematic Review and Meta-Analysis. J Med Internet Res 27:e69639. https://doi.org/10.2196/69639 Fitzpatrick KK, Darcy, Alison, Vierhile M (2017a) Delivering cognitive behavior therapy to young adults with symptoms of depression and anxiety using a fully automated conversational agent (Woebot): a randomized controlled trial. JMIR mental health 4(2):e7785. https://doi.org/10.2196/mental.7785 Fulmer R et al (2018) Using Psychological Artificial Intelligence (Tess) to Relieve Symptoms of Depression and Anxiety: Randomized Controlled Trial. JMIR Ment Health 5(4):e64. https://doi.org/10.2196/mental.9782 Gabrielli S et al (2021) Engagement and Effectiveness of a Healthy-Coping Intervention via Chatbot for University Students During the COVID-19 Pandemic: Mixed Methods Proof-of-Concept Study. JMIR Mhealth Uhealth 9(5):e27965. https://doi.org/10.2196/27965 Gao Y (2025) The impact and application of artificial intelligence technology on mental health counseling services for college students. J Comput Methods Sci Eng 25(2):1686–1701. https://doi.org/10.1177/14727978241302641 Gbollie E, Fayiah et al (2023) Intention to use digital mental health solutions: A cross-sectional survey of university students attitudes and perceptions toward online therapy, mental health apps, and chatbots. Digit HEALTH 9:20552076231216559. https://doi.org/10.1177/20552076231216559 Grové C (2021) Co-developing a Mental Health and Wellbeing Chatbot With and for Young People. Front Psychiatry 11–2020. https://doi.org/10.3389/fpsyt.2020.606041 Harris BR, Maher BM, Wentworth L (2022) Optimizing Efforts to Promote Mental Health on College and University Campuses: Recommendations to Facilitate Usage of Services, Resources, and Supports. J Behav Health Serv Res 49(2):252–258. https://doi.org/10.1007/s11414-021-09780-2 He Y et al (2022) Mental health chatbot for young adults with depressive symptoms during the COVID-19 pandemic: single-blind, three-arm randomized controlled trial. J Med Internet Res 24(11):e40719. https://doi.org/10.2196/40719 Hoagland A, Kipping S (2024) Challenges in promoting health equity and reducing disparities in access across new and established technologies. Can J Cardiol 40(6):1154–1167. https://doi.org/10.1016/j.cjca.2024.02.014 Hopman K, Richards D, Norberg MM (2023) A Digital Coach to Promote Emotion Regulation Skills. Multimodal Technol Interact 7(6):57. https://doi.org/10.3390/mti7060057 Hyseni Duraku, Zamira, Davis H, Hamiti E (2023) Mental health, study skills, social support, and barriers to seeking psychological help among university students: a call for mental health support in higher education. Front Public Health 11:1220614. https://doi.org/10.3389/fpubh.2023.1220614 Jones WH (1985) 'The psychology of loneliness: Some personality issues in the study of social support', Social support: Theory, research and applications (Springer), 225 – 41. https://doi.org/10.1007/978-94-009-5115-0_13 Karyotaki E et al (2020) Sources of stress and their associations with mental disorders among college students: results of the world health organization world mental health surveys international college student initiative. Front Psychol 11:1759. https://doi.org/10.3389/fpsyg.2020.01759 Kavakli M, Li M, Rudra T (2012) Towards the Development of A Virtual Counselor to Tackle Students' Exam Stress. J Integr Des Process Sci 16(1):5–26. https://doi.org/10.3233/jid-2012-0005 Khan N et al (2024) Global insights and the impact of generative AI-ChatGPT on multidisciplinary: a systematic review and bibliometric analysis. Connection Sci 36(1):2353630. https://doi.org/10.1080/09540091.2024.2353630 Klos M, Carolina et al (2021) Artificial Intelligence–Based Chatbot for Anxiety and Depression in University Students: Pilot Randomized Controlled Trial. JMIR Form Res 5(8):e20678. https://doi.org/10.2196/20678 Krauss A (2021) Assessing the Overall Validity of Randomised Controlled Trials. Int Stud Philos Sci 34(3):159–182. https://doi.org/10.1080/02698595.2021.2002676 Kroenke K, Spitzer RL, Williams, Janet BW (2001) The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med 16(9) 606 – 13. https://doi.org/10.1046/j.1525-1497.2001.016009606.x Laestadius L et al (2022) Too human and not human enough: A grounded theory analysis of mental health harms from emotional dependence on the social chatbot Replika. New Media Soc 26(10):5923–5941. https://doi.org/10.1177/14614448221142007 Lattie EG et al (2022) Uptake and effectiveness of a self-guided mobile app platform for college student mental health. Internet interventions 27:100493. https://doi.org/10.1016/j.invent.2021.100493 Li H et al (2023) Systematic review and meta-analysis of AI-based conversational agents for promoting mental health and well-being. NPJ Digit Med 6(1):236. https://doi.org/10.1038/s41746-023-00979-5 Li Y et al (2025) Chatbot-Based Mindfulness-Based Stress Reduction Program for University Students With Depressive Symptoms: Intervention Development and Pilot Evaluation. J Am Psychiatr Nurses Assoc 31(4):398–411. https://doi.org/10.1177/10783903241302092 Liu H et al (2022) Using AI chatbots to provide self-help depression interventions for university students: A randomized trial of effectiveness. Internet Interv 27:100495. https://doi.org/10.1016/j.invent.2022.100495 Liu Z et al (2015) Visualizing the intellectual structure and evolution of innovation systems research: a bibliometric analysis. Scientometrics 103:135–158. https://doi.org/10.1007/s11192-014-1517-y Maciejewski J, Smoktunowicz E (2023) Low-effort internet intervention to reduce students' stress delivered with Meta's Messenger chatbot (Stressbot): A randomized controlled trial. Internet Interventions 33:100653. https://doi.org/10.1016/j.invent.2023.100653 Manole A et al (2024) 'Harnessing AI in Anxiety Management: A Chatbot-Based Intervention for Personalized Mental Health Support', Information , 15 (12) Martinez-Martin N (2023) Viewing CAI as a Tool Within the Mental Health Care System. Am J Bioeth 23(5):57–59. https://doi.org/10.1080/15265161.2023.2191058 Marwaha EB et al (2025) The Digital Bridge: using Digital Intervention to Understand Psychosocial Correlation between COVID-19 Stress and Sleep Quality in Undergraduate Students Before and After COVID-19. Open Psychol J 18. https://doi.org/10.2174/0118743501364965250220093932 Marzi G et al (2025) Guidelines for Bibliometric-Systematic Literature Reviews: 10 steps to combine analysis, synthesis and theory development. Int J Manage Reviews 27(1):81–103. https://doi.org/10.1111/ijmr.12357 Mhlanga D (2023) 'The power of smart technologies for global partnership for development', FinTech and artificial intelligence for sustainable development: The role of smart technologies in achieving development goals (Springer), 291–315. https://doi.org/10.1007/978-3-031-37776-1_13 Mofatteh M (2021) Risk factors associated with stress, anxiety, and depression among university undergraduate students. AIMS Public Health 8(1):36–65. https://doi.org/10.3934/publichealth.2021004 Moldt JA et al (2022) Assessing medical students' perceived stress levels by comparing a chatbot-based approach to the Perceived Stress Questionnaire (PSQ20) in a mixed-methods study. Digit Health 8:20552076221139092 Nadarzynski T et al (2024) Achieving health equity through conversational AI: A roadmap for design and implementation of inclusive chatbots in healthcare. PLOS Digit Health 3(5):e0000492. https://doi.org/10.1371/journal.pdig.0000492 Nations U (2015) 'Transforming our world: The 2030 Agenda for Sustainable Development'. https://sdgs.un.org/2030agenda Nelekar S et al (2022) Effectiveness of embodied conversational agents for managing academic stress at an Indian University (ARU) during COVID-19. Br J Edu Technol 53(3):491–511. https://doi.org/10.1111/bjet.13174 Oghenekaro LU, Okoro CO (2024) Artificial intelligence-based chatbot for student mental health support. Open Access Libr J 11(5):1–14. https://doi.org/10.4236/oalib.1111511 Olla P et al (2025) Deploying a Mental Health Chatbot in Higher Education: The Development and Evaluation of Luna, an AI-Based Mental Health Support System. Computers 14(6):227. https://doi.org/10.3390/computers14060227 Omarov B, Narynov S, Zhumanov Z (2023) Artificial Intelligence-Enabled Chatbots in Mental Health: A Systematic Review. Computers Mater Continua 74(3):5105–5122. https://doi.org/10.32604/cmc.2023.034655 Park SH et al (2019) Designing a Chatbot for a Brief Motivational Interview on Stress Management: Qualitative Case Study. J Med Internet Res 21(4):e12231. https://doi.org/10.2196/12231 Potts C et al (2021a) Correction to: Chatbots to Support Mental Wellbeing of People Living in Rural Areas: Can User Groups Contribute to Co-design? J Technol Behav Sci 6(4):666–666. https://doi.org/10.1007/s41347-021-00222-6 Priestley M et al (2022) Student perspectives on improving mental health support services at university. Counselling Psychother Res 22(1). https://doi.org/10.1002/capr.12391 Salamanca-Sanabria A et al (2023) Exploring the Perceptions of mHealth Interventions for the Prevention of Common Mental Disorders in University Students in Singapore: Qualitative Study. J Med Internet Res 25:e44542. https://doi.org/10.2196/44542 Sebastian J, Richards D (2017) Changing stigmatizing attitudes to mental health via education and contact with embodied conversational agents. Comput Hum Behav 73:479–488. https://doi.org/10.1016/j.chb.2017.03.071 Syed SA (2024) The role of AI in alleviating loneliness among adults in the United States. Int J Eng Technol Res Manage (IJETRM) 8(04):404–421 Trappey AJC et al (2022) 'Development of an Empathy-Centric Counseling Chatbot System Capable of Sentimental Dialogue Analysis', Processes , 10 (5). https://doi.org/10.3390/pr10050930 Vaidyam A, Nrusimha et al (2019) Chatbots and conversational agents in mental health: a review of the psychiatric landscape. Can J Psychiatry 64(7):456–464. https://doi.org/10.1177/0706743719828977 Vereschagin M et al (2024) Effectiveness of the Minder Mobile Mental Health and Substance Use Intervention for University Students: Randomized Controlled Trial. J Med Internet Res 26:e54287. https://doi.org/10.2196/54287 Vilaza GN, McCashin D (2021) Is the Automation of Digital Mental Health Ethical? Applying an Ethical Framework to Chatbots for Cognitive Behaviour Therapy. Front Digit Health 3–2021. https://doi.org/10.3389/fdgth.2021.689736 Zhai C, Wibowo S (2022) 'A systematic review on cross-culture, humor and empathy dimensions in conversational chatbots: the case of second language acquisition', Heliyon , 8 (12). https://doi.org/10.1016/j.heliyon.2022.e12056 Zhai S et al (2025) Technology-driven support: exploring the impact of artificial intelligence on mental health in higher education. Educ Inform Technol 30(10):13931–13959. https://doi.org/10.1007/s10639-025-13352-8 Zhou X et al (2021) Are online mental health interventions for youth effective? A systematic review. J Telemed Telecare 27(10):638–666. https://doi.org/10.1177/1357633X211047285 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 07 Mar, 2026 Reviews received at journal 17 Feb, 2026 Reviews received at journal 13 Feb, 2026 Reviewers agreed at journal 12 Feb, 2026 Reviewers agreed at journal 12 Feb, 2026 Reviews received at journal 11 Feb, 2026 Reviewers agreed at journal 11 Feb, 2026 Reviewers agreed at journal 10 Feb, 2026 Reviewers invited by journal 10 Feb, 2026 Editor assigned by journal 09 Oct, 2025 Editor invited by journal 27 Sep, 2025 Submission checks completed at journal 20 Sep, 2025 First submitted to journal 19 Sep, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7486943","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Systematic Review","associatedPublications":[],"authors":[{"id":591181172,"identity":"c2a32793-41db-42ab-bb42-d0179b2b6f79","order_by":0,"name":"Xing Yi","email":"","orcid":"","institution":"University of Malaya","correspondingAuthor":false,"prefix":"","firstName":"Xing","middleName":"","lastName":"Yi","suffix":""},{"id":591181181,"identity":"f9e95008-be00-4b68-a098-ecd38bf63491","order_by":1,"name":"Ying Chen","email":"","orcid":"","institution":"University of Malaya","correspondingAuthor":false,"prefix":"","firstName":"Ying","middleName":"","lastName":"Chen","suffix":""},{"id":591181183,"identity":"a9fc68f8-1ea0-4c5b-898e-86c4d236ba97","order_by":2,"name":"Sau Cheong Loh","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAnUlEQVRIiWNgGAWjYBACPgYeBgbGBhswR4IoLWwQLWmkazlMihb23sMffu44n7i2gfngbR6itPCcS5PsPXM7cdsBtmRr4rRI5JgxM7aBtPCYSROrxfgzY9s5oBb+b0RrMZBmbDsAsoWNSC08Z8wke9uSjbcdZjO2nEOMFn72HuMPP9vsZLcdb3544w0xWhCAmTTlo2AUjIJRMArwAQC4oy3MGp9d9AAAAABJRU5ErkJggg==","orcid":"","institution":"University of Malaya","correspondingAuthor":true,"prefix":"","firstName":"Sau","middleName":"Cheong","lastName":"Loh","suffix":""},{"id":591181194,"identity":"596c8d01-3a92-4955-96cc-af31468f597a","order_by":3,"name":"Yan-Li Siaw","email":"","orcid":"","institution":"University of Malaya","correspondingAuthor":false,"prefix":"","firstName":"Yan-Li","middleName":"","lastName":"Siaw","suffix":""},{"id":591181204,"identity":"e223d49e-0462-40c5-bfb4-8dc45d9c79e1","order_by":4,"name":"Li Jie Hao","email":"","orcid":"","institution":"Zhejiang Wanli University","correspondingAuthor":false,"prefix":"","firstName":"Li","middleName":"Jie","lastName":"Hao","suffix":""},{"id":591181206,"identity":"4c5064cf-f043-4171-b236-39fe789d61ce","order_by":5,"name":"Hassan Abuhassna","email":"","orcid":"","institution":"Sunway University","correspondingAuthor":false,"prefix":"","firstName":"Hassan","middleName":"","lastName":"Abuhassna","suffix":""}],"badges":[],"createdAt":"2025-08-29 09:08:31","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7486943/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7486943/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":102744032,"identity":"8daf6929-e068-4dcd-9751-ed0f6ce0a622","added_by":"auto","created_at":"2026-02-16 08:26:14","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":364121,"visible":true,"origin":"","legend":"\u003cp\u003eFlow chart of the search process through PRISMA.\u003c/p\u003e","description":"","filename":"Fig.1FlowchartofthesearchprocessthroughPRISMA.png","url":"https://assets-eu.researchsquare.com/files/rs-7486943/v1/284bbb1547d42ce0c5d381be.png"},{"id":102743999,"identity":"0eb35957-6c44-447f-a287-898e689f6d5c","added_by":"auto","created_at":"2026-02-16 08:26:07","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":73011,"visible":true,"origin":"","legend":"\u003cp\u003eAnnual scientific production on AI-based chatbots for mental health support in university settings from 2012 to 2025.\u003c/p\u003e","description":"","filename":"Fig.2.NumberofglobalpublicationsonAIbasedchatbotsinmentalhealthsupportforuniversitystudents..png","url":"https://assets-eu.researchsquare.com/files/rs-7486943/v1/d20a87518f39854ca641a7e8.png"},{"id":102744125,"identity":"84bab0ba-62a4-4ee8-b2a5-135176ae5fe7","added_by":"auto","created_at":"2026-02-16 08:26:37","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":145203,"visible":true,"origin":"","legend":"\u003cp\u003eCountry collaboration map showing the international cooperation in publications related to AI-based mental health chatbots.\u003c/p\u003e","description":"","filename":"Fig.3.CountryCollaborationMap.png","url":"https://assets-eu.researchsquare.com/files/rs-7486943/v1/e267276c6c9b099eab20adf6.png"},{"id":102744029,"identity":"20e68e2b-79fd-49c2-9b0b-b300e339030e","added_by":"auto","created_at":"2026-02-16 08:26:13","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":233250,"visible":true,"origin":"","legend":"\u003cp\u003eCo-citation network of authors in the field of AI-based chatbots and student mental health research.\u003c/p\u003e","description":"","filename":"Fig.4.Cocitationnetwork.png","url":"https://assets-eu.researchsquare.com/files/rs-7486943/v1/377ca5d1d4e9501e2a620231.png"},{"id":102744028,"identity":"4b19712b-8f75-4c56-b35d-9a2a272f35f4","added_by":"auto","created_at":"2026-02-16 08:26:13","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":402085,"visible":true,"origin":"","legend":"\u003cp\u003eWord cloud of the most frequent keywords in the literature on AI chatbots and student mental health.\u003c/p\u003e","description":"","filename":"Fig.5.Wordcloud.png","url":"https://assets-eu.researchsquare.com/files/rs-7486943/v1/7012ca87264e41c39d9b9b03.png"},{"id":102743997,"identity":"3628be7a-ab20-431b-8e34-8c2bad5ba6a5","added_by":"auto","created_at":"2026-02-16 08:26:06","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":50300,"visible":true,"origin":"","legend":"\u003cp\u003eTrend topics in AI-based chatbot research for student mental health, illustrating term usage by year and frequency.\u003c/p\u003e","description":"","filename":"Fig.6.Trendtopics.png","url":"https://assets-eu.researchsquare.com/files/rs-7486943/v1/af09cb3fbb61da8a2e7ddac9.png"},{"id":102743998,"identity":"5e08915b-4cb8-4f71-90db-84c41e3a68ca","added_by":"auto","created_at":"2026-02-16 08:26:07","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":623776,"visible":true,"origin":"","legend":"\u003cp\u003eCo-occurrence network of keywords in studies on AI-based chatbots and mental health among university students.\u003c/p\u003e","description":"","filename":"Fig.7.cooccurrences.png","url":"https://assets-eu.researchsquare.com/files/rs-7486943/v1/a388427ab68c2651022bcdfd.png"},{"id":102749321,"identity":"6fdf2474-d9b9-45fa-b990-8000cc4427eb","added_by":"auto","created_at":"2026-02-16 09:12:23","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2996397,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7486943/v1/ad40c0b4-ad78-42a5-beb3-2bb65e1ba77c.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"AI Chatbots for University Student Mental Health: Bibliometric Mapping and Systematic Review with Future Directions for Sustainable Development Goals","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eUniversity students are increasingly experiencing psychological distress due to academic demands, career uncertainty, and social challenges (Karyotaki et al. \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Mofatteh \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). As they represent the next generation of societal leaders, safeguarding their mental health and well-being is both an educational priority and a broader public health concern. It not only supports their academic success but also contributes to global aspirations for healthier, more inclusive, and equitable societies, as reflected in international frameworks in United Nations\u0026rsquo; 2030 Agenda for Sustainable Development Goals (SDG), such as SDG 3 for Good Health and Well-being, SDG 4 for Quality Education, and SDG 10 for Reduced Inequalities (Nations \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2015\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAlthough universities have introduced initiatives such as mental health courses and counseling services, these resources are often overwhelmed, difficult to scale, and lack personalization (Priestley et al. \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). This creates disparities in access to timely support, particularly for students from underserved backgrounds. Addressing these gaps requires innovative, scalable solutions that can complement traditional services while maintaining quality and equity.\u003c/p\u003e \u003cp\u003eIn recent years, AI-based chatbots have emerged as promising tools for mental health intervention, offering continuous availability, cost-effectiveness, and user anonymity. Advances in generative AI (e.g., GPT-3, GPT-4) have enabled chatbots to provide empathetic, context-aware dialogue (Bansal et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), making them attractive complements to traditional counseling, especially for university student populations (H. Liu et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Manole et al. \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). While existing studies have explored chatbot effectiveness for various mental health outcomes, they often focus on specific applications or short-term results. There remains a lack of comprehensive analysis that maps the intellectual structure of this field, identify influential contributors, and reveal thematic trends. Long-term considerations, such as ethical risks and sustainable development, also require further attention.\u003c/p\u003e \u003cp\u003eTherefore, this study explicitly positions itself within the pursuit of the SDGs and integrates bibliometric methods with a systematic literature review. The research objectives are to chart publication trends, identify key contributors (journals, countries, institutions, landmark papers), trace thematic evolution over time, and uncover the future directions. The specific research questions are:\u003c/p\u003e \u003cp\u003e1.Publication trends: How have publications on AI chatbots for university student mental health been distributed over time?\u003c/p\u003e\n\u003ch3\u003e2.Key contributors: Which journals, countries, institutions, and papers have been most influential in this field?\u003c/h3\u003e\n\n\u003ch3\u003e3.Thematic trends: What are the key words, prevailing research topics, and how have they evolved?\u003c/h3\u003e\n\u003cp\u003e4.Future directions: What are suggested directions for future research in AI-based student mental health support related to SDG?\u003c/p\u003e"},{"header":"2. Literature Review","content":"\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.1 AI-based chatbots and mental health potentials\u003c/h2\u003e \u003cp\u003eAI-powered chatbots are computer programs that use natural language processing to simulate human conversation. In recent years, particularly transformer-based generative models (GPT-3, GPT-4), have greatly enhanced chatbots\u0026rsquo; abilities to mimic human empathy and responsiveness (Adamopoulou and Moussiades 2020; Bansal et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). These advanced models produce adaptive, nuanced replies, improving the quality of conversational support (Bansal et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). By leveraging sentiment analysis, chatbots can recognize users\u0026rsquo; emotion and respond with empathic understanding, which can facilitate emotional disclosure and reduce perceived stigma (Denecke and Gabarron \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). This reduction in stigma is crucial for achieving equitable access to mental health support, which is aligned with the aim of SDG 10 to reduce various forms of inequality (Nations \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). From a psychological standpoint, such empathic engagement aligns with social support theory (Jones \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e1985\u003c/span\u003e): when users feel heard and understood, stress and loneliness decrease (Syed \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Chatbots can also deliver structured interventions, such as cognitive behavioral therapy (CBT) or exercises and guided self-reflection (Eltahawy et al. \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Farzan et al. \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Their potential to offer scalable, basic mental health support significantly contributes to Sustainable Development Goal 3 and 4. This target aims to \"reduce by one third premature mortality from non-communicable diseases through prevention and treatment and promote mental health and well-being\" by 2030 (Nations \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2015\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eDespite this potential, most research to date has focused on the technical development of chatbots or isolated case studies. Few studies integrate technical and psychological perspectives, especially within the higher education context. In other words, literature often examines chatbot algorithms or short-term efficacy, without synthesizing broader trends or combining insights from computer science and mental health disciplines.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.2 AI-based chatbot in supporting university students\u0026rsquo; mental health\u003c/h2\u003e \u003cp\u003eUniversity students are generally considered a high-risk group for stress and mental illness (Karyotaki et al. \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). To support student well-being, many higher education institutes offer mental health courses and free counseling, but demand often exceeds capacity (Hyseni Duraku et al. 2023). In this setting, chatbots have significantly expanded access to mental health resources. They provide available, scalable, and low-cost support, which can lower barriers for students facing financial or stigma-related obstacles (Manole et al. \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Omarov et al. \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Consequently, chatbots have become a popular option for students seeking help (H. Liu et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). By offering immediate guidance and coping strategies, these tools can contribute to early intervention, emotional regulation, and overall well-being (Klos et al. \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; H. Liu et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). AI chatbot contributes to their academic success and overall well-being, core aspects of SDG 4 \"ensure inclusive and equitable quality education\"(Nations \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2015\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eRecent studies highlight the promise of generative AI chatbots for student mental health (Klos et al. \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; H. Liu et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). However, most focus on singular issues (e.g., reducing anxiety or stress) without providing a holistic view of student mental health. Existing meta-analyses and reviews have confirmed chatbots\u0026rsquo; effectiveness for youth mental health (Feng et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; H. Li et al. \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) or categorized chatbot interventions by type and approach (Omarov et al. \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Yet many of these analyses do not specifically address the higher education context. Students\u0026rsquo; psychological needs are often more complex and multifaceted, influenced by academic pressures and transitional life stages. Previous literature lacks a broad mapping of the field of AI chatbots for university student mental health.\u003c/p\u003e \u003cp\u003eSeveral gaps remain. First, there is a lack of bibliometric analyses that systematically uncover the intellectual structure of the field, trace the evolution of knowledge, and identify influential journals, seminal works, leading institutions, and contributing countries. Second, few studies have combined bibliometric mapping with qualitative synthesis to bridge technical, psychological, and educational perspectives on the university student sample. Finally, although AI-based chatbots offer accessible and scalable mental health support for university students, important long-term considerations, such as ethical challenges (Chan \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), and potential development in this field should be taken into consideration. As these tools become increasingly integrated into student life, moving beyond short-term evaluations and exploring future directions with a focus on sustainability, equity, and long-term well-being outcomes (Nations \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) is essential. There is a need for systematic bibliometric analysis to uncover the structure of research output and for qualitative synthesis to identify overarching gaps and future research pathways in this interdisciplinary domain.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Method","content":"\u003cp\u003eThis study employs a dual-method approach integrating bibliometric analysis and systematic literature review (SLR). Bibliometric analysis systematically maps knowledge domains and citation networks, revealing publication/citation trends, key topics, thematic evolution, and influential sources, countries, institutions, and documents (Donthu et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Z. Liu et al. \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). SLR identifies future directions and provides qualitative depth. These methods are mutually reinforcing, and their integration is increasingly advocated for robust research synthesis (Marzi et al. \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), as demonstrated in prior studies (Abuhassna et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Search strategies\u003c/h2\u003e \u003cp\u003eWe searched the Web of Science (WoS) Core Collection and Scopus databases for peer-reviewed English-language articles up to 5 July, 2025. Search terms combined AI/chatbot-related keywords (e.g., \u0026ldquo;AI chatbot,\u0026rdquo; \u0026ldquo;conversational agent,\u0026rdquo; \u0026ldquo;GPT-4\u0026rdquo;), mental health terms (e.g., \u0026ldquo;mental health,\u0026rdquo; \u0026ldquo;well-being,\u0026rdquo; \u0026ldquo;depression,\u0026rdquo; \u0026ldquo;anxiety,\u0026rdquo; \u0026ldquo;stress\u0026rdquo;), and student context terms (e.g., \u0026ldquo;university student,\u0026rdquo; \u0026ldquo;college student,\u0026rdquo; \u0026ldquo;higher education,\u0026rdquo; \u0026ldquo;undergraduate\u0026rdquo;). Boolean operators (AND/OR) and wildcards were used to capture variations. The initial search yielded 417 records (148 from WoS, 269 from Scopus). We excluded duplicates, non-article document types (reviews, conference papers, etc.), and non-English works. We then screened titles, abstracts, and keywords for relevance to the intersection of AI/chatbots, mental health, and university students. This process (see PRISMA flow diagram in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) resulted in 58 articles for bibliometric analysis. From these, we identified the 20 most-cited empirical papers (excluding reviews) as the basis for our systematic literature review.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Data analysis\u003c/h2\u003e \u003cp\u003eFor bibliometric analysis, we used the R bibliometric package, Biblioshiny and VOSviewer software. We examined annual publication counts, author and country collaborations, and distribution of publications by source, country, institution, and document. We identified the most cited documents and conducted co-citation analysis to map the intellectual base. Keyword co-occurrence analysis and trend-topic analysis were performed to detect major themes and their evolution.\u003c/p\u003e \u003cp\u003eFor the systematic review, we categorized the top 20 cited papers into thematic clusters (derived from keyword co-occurrence). For each paper, we extracted information on research design, objectives, key findings, limitations, and proposed future directions. Synthesizing these data across clusters, we identified the recommended future research directions.\u003c/p\u003e \u003c/div\u003e"},{"header":"4. Results","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e4.1 Global publication trends\u003c/h2\u003e \u003cp\u003eWe identified 58 relevant articles published between 2012 and July 2025, with an average document age of 1.74 years. Annual output remained very low until 2021 but then grew rapidly (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The first relevant publication appeared in 2012, followed by a hiatus until 2017. From 2018\u0026ndash;2021, only 1\u0026ndash;5 papers appeared annually. Beginning in 2022, publication counts increased sharply: 4 papers in 2022, 8 in 2023, 12 in 2024, and 25 by July 2025. Overall growth accelerated after 2021, reflecting heightened research interest. This trend suggests that AI-based student mental health chatbots are an emerging topic, which may be due to the improvements in AI technology and increased student mental health concerns (e.g., due to the COVID-19 pandemic).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e4.2 Analysis of the most influential and productive agents (countries, affiliations, sources and documents)\u003c/h2\u003e \u003cp\u003eCountries and Institutions: Thirty-three countries contributed to the 58 articles. The result was based on co-authorship data, as publications often involve international collaborations, the sum of country contributions exceeds the total number of articles China was listed as a contributing country in 46 of the 58 publications, followed by South Korea (19), the USA (19), Canada (14), Australia (12), Jordan (12), and the UK (11) (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Notably, countries with strong publication counts include both traditional research powers and emerging academic communities (e.g., South Africa, Saudi Arabia, Lebanon with 6\u0026ndash;10 each). International collaboration networks (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e) show strong links between the U.S. and Canada and within the Middle East, as well as connections among China, Australia, and the UK. Nevertheless, the overall international co-authorship rate was only 25.4%, indicating space for broader collaboration.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eTop 10 countries and institutions for AI\u0026ndash;based chatbots in mental support among university students.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRank\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCountry\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCounts\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e \u003cp\u003eAffiliation\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eCounts\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCHINA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e46\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003eThe University of British Columbia\u003c/em\u003e(Canada)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSOUTH KOREA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003eMacquarie University\u003c/em\u003e(Australia)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUSA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eThe University of Jordan (Jordan)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCANADA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003eMing Chuan University\u003c/em\u003e(Taiwan,China)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAUSTRALIA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eUniversity of Huddersfield (UK)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eJORDAN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eHanyang University (South Korea)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUK\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003eHubei University of Chinese Medicine\u003c/em\u003e (China)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSOUTH AFRICA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eImperial College London (UK)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSAUDI ARABIA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNational Tsing Hua University (Taiwan,China)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLEBANON\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSeoul National University (South Korea)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe University of British Columbia (Canada) was the most productive institution (10 papers), followed by Macquarie University (Australia, 8) and the University of Jordan (7). Other notable institutions (4\u0026ndash;5 papers each) were primarily in East Asia (e.g., Ming Chuan University, Hanyang University, Hubei University of Chinese Medicine, National Tsing Hua University, Seoul National University) and the UK (University of Huddersfield, Imperial College London). The top institutions span North America, Asia, Australia, the UK, and the Middle East, reflecting growing global interest, more details can be found in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003eJournals: The 58 papers appeared in 46 different journals. The \u003cem\u003eJournal of Medical Internet Research\u003c/em\u003e (JMIR) published the most (4 papers), followed by \u003cem\u003eDigital Health\u003c/em\u003e and \u003cem\u003eJMIR Mental Health\u003c/em\u003e (3 each). Among top journals, \u003cem\u003eJMIR MHealth\u003c/em\u003e and \u003cem\u003eUHealth\u003c/em\u003e had the highest impact factor (6.1), followed by \u003cem\u003eJMIR\u003c/em\u003e (6.0) and \u003cem\u003eJMIR Mental Health\u003c/em\u003e (5.8). Most top outlets were in Q1, indicating high source quality (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eTop 10 journals in publications on AI\u0026ndash;based chatbots related to mental support among university students.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRank\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eJournal\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCount\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eIF\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eQuartile\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eJOURNAL OF MEDICAL INTERNET RESEARCH\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDIGITAL HEALTH\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eJMIR MENTAL HEALTH\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eINTERNATIONAL JOURNAL OF HUMAN-COMPUTER INTERACTION\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eINTERNET INTERVENTIONS-THE APPLICATION OF INFORMATION TECHNOLOGY IN MENTAL AND BEHAVIOURAL HEALTH\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eJMIR FORMATIVE RESEARCH\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eJMIR MHEALTH AND UHEALTH\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eJOURNAL OF AFFECTIVE DISORDERS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eACM TRANSACTIONS ON APPLIED PERCEPTION\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAPPLIED SCIENCES-BASEL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eKey Documents: The most-cited article was \u003cem\u003e\u0026ldquo;Delivering Cognitive Behavior Therapy to Young Adults With Symptoms of Depression and Anxiety Using a Fully Automated Conversational Agent (Woebot): A Randomized Controlled Trial\u0026rdquo;\u003c/em\u003e by Fitzpatrick et al. (2017b) with 1,012 citations. This landmark Woebot RCT demonstrated feasibility and preliminary efficacy of an automated CBT chatbot for college students. It was followed by Fulmer et al. (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) (\u003cem\u003eUsing Psychological AI (Tess)\u003c/em\u003e, 284 cites), which showed feasibility of an AI chatbot to reduce depression/anxiety symptoms, and Park et al. (\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) (\u003cem\u003eDesigning a Chatbot...Brief Motivational Interview\u003c/em\u003e), with 99 cites (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). These top studies established foundational evidence in the field. We also analyzed the co-citation network, Fitzpatrick et al. (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2017a\u003c/span\u003e), Kroenke et al. (\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2001\u003c/span\u003e), Vaidyam et al. (\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) and Auerbach et al. (2018). These articles in the co-citation network (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e) laid the foundation for the development of the research field on AI-based chatbots in support of mental health among university students.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eTop 10 most cited documents\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePaper\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDOI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTotal Citations\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTC per Year\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNormalized TC\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e(Fitzpatrick et al. 2017b)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2196/mental.7785\u003c/span\u003e\u003cspan address=\"10.2196/mental.7785\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1012\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e112.44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.89\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e(Fulmer et al. \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2018\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2196/mental.9782\u003c/span\u003e\u003cspan address=\"10.2196/mental.9782\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e284\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e35.50\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e(Park et al. \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2019\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2196/12231\u003c/span\u003e\u003cspan address=\"10.2196/12231\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e14.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e(H. Liu et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2022\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.invent.2022.100495\u003c/span\u003e\u003cspan address=\"10.1016/j.invent.2022.100495\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e23.75\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2.77\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e(Gabrielli et al. \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2021\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2196/27965\u003c/span\u003e\u003cspan address=\"10.2196/27965\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e81\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e16.20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.56\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e(Abdaljaleel et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2024\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41598-024-52549-8\u003c/span\u003e\u003cspan address=\"10.1038/s41598-024-52549-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e70\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e35.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e5.45\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e(Zhou et al. \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2021\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1177/1357633X211047285\u003c/span\u003e\u003cspan address=\"10.1177/1357633X211047285\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e66\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e13.20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.27\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e(Klos et al. \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2021\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2196/20678\u003c/span\u003e\u003cspan address=\"10.2196/20678\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e65\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e13.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.25\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e(Sebastian and Richards \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2017\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.chb.2017.03.071\u003c/span\u003e\u003cspan address=\"10.1016/j.chb.2017.03.071\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6.44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.11\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e(Potts et al. 2021b)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s41347-021-00222-6\u003c/span\u003e\u003cspan address=\"10.1007/s41347-021-00222-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e31\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.6\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003e4.3 Key words and trend topics\u003c/b\u003e \u003c/p\u003e \u003cp\u003eThe most frequent keywords included \u0026ldquo;ChatGPT,\u0026rdquo; \u0026ldquo;AI,\u0026rdquo; \u0026ldquo;student anxiety,\u0026rdquo; \u0026ldquo;depression,\u0026rdquo; \u0026ldquo;stress,\u0026rdquo; and \u0026ldquo;mental health\u0026rdquo;, as revealed by the word cloud visualization (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). Our trend topic analysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e) shows that \u0026ldquo;conversational agents\u0026rdquo; has been a consistent focus throughout the period. Notably, terms like \u0026ldquo;AI\u0026rdquo; and \u0026ldquo;ChatGPT\u0026rdquo; became prominent around 2023, reflecting the impact of generative AI breakthroughs (e.g., ChatGPT release in late 2022). The emergence of \u0026ldquo;mHealth\u0026rdquo; as a trending topic in 2023 indicates growing integration with mobile technologies. Terms \u0026ldquo;female\u0026rdquo; and \u0026ldquo;student\u0026rdquo; appearing since 2021 suggest increasing attention to gender-specific aspects. \u0026ldquo;Stress\u0026rdquo; began trending in 2022, highlighting heightened interest in using chatbots for stress reduction.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eCo-occurrence analysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e) identified 56 items meeting the threshold of at least 2 co-occurrences. The network visualization reveals keyword clusters, each representing distinct thematic groups, where larger font sizes denote greater prominence. Of these, 57 items were grouped into 5 clusters. Across all clusters, keywords such as \u003cem\u003echatbot\u003c/em\u003e, \u003cem\u003emental health\u003c/em\u003e, \u003cem\u003eacademic stress\u003c/em\u003e, \u003cem\u003eanxiety\u003c/em\u003e, \u003cem\u003eAI\u003c/em\u003e, depression, and university student emerged as highly salient. The interconnections within each cluster further delineate subthemes under the broader research focus. We identified the top plus key words under each cluster and considered the high-frequency words and the content of the individual articles to find potential subtopics for further systematic review analysis. The contents of the five clusters are presented in Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e. Cluster 1 emphasized the AI chatbot for university students\u0026rsquo; academic stress; cluster 2 was labeled for chatbot prevention or treatment, such as depression and anxiety symptoms, in the high-risk group; cluster 3 highlighted chatbots for self-help with emotional or mental health issues; cluster 4 focused on evaluations of chatbot intervention; and cluster 5 focused on the user experience or perceptions of chatbots.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSub-themes clusters\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCluster #\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTop keyword plus\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePotential topic\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAcademic stress, achievement, stress management, performance\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eChatbot in stress setting\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMental health, mental disorder, illness, symptoms, anxiety, depression\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eChatbot for high-risk group\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ephq-9, loneliness, self-disclosure, emotion, self-help, well-being,\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eChatbot for emotion support\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003erandomized controlled trial, questionnaire, validation, scale, intervention, effectiveness\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eResearch design and intervention evaluation\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUser experience, conversational agent, human computer interaction, use intension\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eChatbot technology and user experience\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e4.4 Future directions based on a systematic literature review\u003c/h2\u003e \u003cp\u003eAfter confirming the subthemes, we conducted a systematic review and investigated the top 4 cited articles under each cluster in terms of their method, objective and future directions. The outlines of these 20 articles are presented in Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e. Future directions are highlighted below.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOutline of the systematic literature review\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCluster\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eArticle\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMethod\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eObjective(s)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFuture Direction\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(Nelekar et al. \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) Nelekar et al., \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2022\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eQuasi experimental\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eEvaluate\u0026nbsp;the ECA\u0026rsquo;s effectiveness in reducing academic stress among Indian students.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1. Longitudinal Studies\u003c/p\u003e \u003cp\u003e2. Larger/Diverse Samples\u003c/p\u003e \u003cp\u003e3. Enhanced Personalization\u003c/p\u003e \u003cp\u003e4. Expanded Content\u003c/p\u003e \u003cp\u003e5. Integration with Institutions\u003c/p\u003e \u003cp\u003e6. Cross-Cultural Validation\u003c/p\u003e \u003cp\u003e7. Behavior Tracking\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(Kavakli et al. \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2012\u003c/span\u003e) Kavakli et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2012\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eQualitative and exploratory\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eExplore how interaction design can be used to create an artificial therapist capable of supporting mental health interventions, with a particular focus on stress management for university students.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1. Develop higher-fidelity prototypes\u003c/p\u003e \u003cp\u003e2. Conduct empirical studies with real users and mental health practitioners\u003c/p\u003e \u003cp\u003e3. Incorporate emotion detection and adaptive feedback\u003c/p\u003e \u003cp\u003e4. Explore long-term user engagement\u003c/p\u003e \u003cp\u003e5. Investigate multimodal interfaces\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(Moldt et al. \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) Moldt et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2022\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003emixed-methods research design\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eEvaluate the effectiveness of a chatbot in assessing the stress levels of medical students(N\u0026thinsp;=\u0026thinsp;284) in everyday conversations and identify the main condition for accepting a chatbot as a conversational partner based on validated stress instruments\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1. Improve Chatbot Technical Performance\u003c/p\u003e \u003cp\u003e2. Develop a User-Centered Bot Design\u003c/p\u003e \u003cp\u003e3. Increase Transparency in Bot Functionality\u003c/p\u003e \u003cp\u003e4. Enhance Psychometric Validity of Measurement\u003c/p\u003e \u003cp\u003e5. Expand Cultural and Linguistic Scope\u003c/p\u003e \u003cp\u003e6. Contribute to AI and Education Research Communities\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(Maciejewski and Smoktunowicz \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) Maciejewski \u0026amp; Smoktunowicz, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2023\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eparallel\u0026nbsp;randomized controlled trial\u0026nbsp;(RCT) with two conditions compared on three time points\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eDetermine whether a psychological self-guided intervention could be effective when delivered with a low-effort approach through a chatbot on Meta's Messenger among 372 Polish university students.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1. Include Active Control Comparisons\u003c/p\u003e \u003cp\u003e2. Enhance Measurement of Engagement\u003c/p\u003e \u003cp\u003e3. Adopt Participatory Design Approaches\u003c/p\u003e \u003cp\u003e4. Explore Multimodal Interaction Features\u003c/p\u003e \u003cp\u003e5. Evaluate AI-Enhanced Chatbots\u003c/p\u003e \u003cp\u003e6. Optimize Intervention Intensity\u003c/p\u003e \u003cp\u003e7. Incorporate Booster Sessions or Follow-ups\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(Fitzpatrick et al. 2017b) Fitzpatrick et al., 2017\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003erandomized controlled trial (RCT) design\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eThe study's goal was to assess the feasibility, acceptability, and preliminary efficacy of a completely automated conversational agent for delivering a self-help program to college students who self-identify as suffering from anxiety and depression.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1. Enhanced Trial Design\u003c/p\u003e \u003cp\u003e2. Engagement Tracking\u003c/p\u003e \u003cp\u003e3. Population Expansion\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(Fulmer et al. \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) Fulmer et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2018\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003erandomized controlled trial experimental study\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAssess the feasibility and usefulness of utilizing Tess, an integrative psychological AI, to reduce self-identified symptoms of depression and anxiety in college students(n\u0026thinsp;=\u0026thinsp;75).\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1. Test in\u0026nbsp;clinical population\u003c/p\u003e \u003cp\u003e2.Compare the control group against Traditional therapy, Teletherapy, VR/interactive courses\u003c/p\u003e \u003cp\u003e3.Integrate\u0026nbsp;multimodal emotion recognition\u003c/p\u003e \u003cp\u003e4.Ethical safeguard\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(Klos et al. \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) Klos et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2021\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003epilot randomized controlled trial\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eassess the viability, acceptability, and possible impact of utilizing Tess, a chatbot, for assessing symptoms of depression and anxiety in university students.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1. larger samples\u003c/p\u003e \u003cp\u003e2. Waitlist controls\u003c/p\u003e \u003cp\u003e3.Stratified/unbalanced randomization\u003c/p\u003e \u003cp\u003e4. Participant diversity\u003c/p\u003e \u003cp\u003e5. Intervention design (Therapy-specific chatbots, Hybrid models)\u003c/p\u003e \u003cp\u003e6. Comparative research\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(Y. Li et al. \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) Li et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2025\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003esingle-group pretest-posttest study\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eevaluate the feasibility, acceptability, safety, and preliminary efficacy of a chatbot-based MBSR intervention for university students with depressive symptoms(N\u0026thinsp;=\u0026thinsp;30)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003elarger, controlled sample, extending the follow-up period, and collecting data on confounding factors to better assess the lasting effects of the intervention\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(H. Liu et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) Liu et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2022\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003erandomized controlled trial\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eThis study compared chatbot therapy to well-established bibliotherapy to gather evidence for chatbot therapy's effectiveness as a convenient, low-cost, interactive self-help intervention for depression\u0026nbsp;(n\u0026thinsp;=\u0026thinsp;52).\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1. Technical enhancements\u003c/p\u003e \u003cp\u003e2. Content expansion\u003c/p\u003e \u003cp\u003e3. Diverse demographics\u003c/p\u003e \u003cp\u003e4. Conduct longitudinal studies\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(Salamanca-Sanabria et al. \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) Salamanca-Sanabria et al., \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2023\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eInterview\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eexamine how university students(N\u0026thinsp;=\u0026thinsp;30) and mental health advocates in Singapore perceive mental health services, awareness campaigns, and mHealth interventions, with particular emphasis on the acceptability of conversational agent technologies for the prevention of anxiety and depression\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1. Targeting holistic wellbeing\u003c/p\u003e \u003cp\u003e2. Adapting content to local and contextual barriers\u003c/p\u003e \u003cp\u003e3. Collaborating with government and local organizations\u003c/p\u003e \u003cp\u003e4. Carefully managing the use of incentives\u003c/p\u003e \u003cp\u003e5. Expanding research across different populations and cultural contexts\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(Hopman et al. \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) Hopman et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2023\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eQuasi experimental/Pilot study/Pre-post design without control group\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eExamines whether an ECA can successfully deliver a highly targeted cognitive emotion regulation psychoeducation intervention(N\u0026thinsp;=\u0026thinsp;138)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1. Implement Controlled Experimental Designs\u003c/p\u003e \u003cp\u003e2. Evaluate Long-Term Impact\u003c/p\u003e \u003cp\u003e3. Test with the Target Clinical Population\u003c/p\u003e \u003cp\u003e4. Enhance inclusivity through Personalization\u003c/p\u003e \u003cp\u003e5. Gather Feedback from High-Rating Users\u003c/p\u003e \u003cp\u003e6. Refine Conversation Design\u003c/p\u003e \u003cp\u003e7. Adapt ERICA for MHealth Delivery\u003c/p\u003e \u003cp\u003e8. Engage Stakeholders in Codesign\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(Potts et al. \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2021a\u003c/span\u003e) Potts et al., 2021b\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eQualitative research using Living Labs methodology\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eThis study aims to codesign a chatbot that supports mental wellbeing among people in rural areas (include university students\u0026rsquo; sample)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1. Advance chatbot capabilities\u003c/p\u003e \u003cp\u003e2. Integrate machine learning techniques\u003c/p\u003e \u003cp\u003e3. Address the challenge of empathetic dialog\u003c/p\u003e \u003cp\u003e4. Develop multidisciplinary frameworks\u003c/p\u003e \u003cp\u003e5. Reimagine chatbot roles as early detection and self-monitoring tools\u003c/p\u003e \u003cp\u003e6. Explore personalization and ethical design\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(Gabrielli et al. \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) Gabrielli et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2021\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMixed Methods Proof-of-Concept Study\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003econduct a proof-of-concept evaluation measuring the engagement and effectiveness of Atena, a psychoeducational chatbot supporting healthy coping with stress and anxiety, among a population of university students(N\u0026thinsp;=\u0026thinsp;71)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1. Rigorous Efficacy Testing\u003c/p\u003e \u003cp\u003e2. Enhanced Engagement Measurement\u003c/p\u003e \u003cp\u003e3. User-Centered Development:\u0026nbsp;Employ user-centered methodology\u003c/p\u003e \u003cp\u003e4. Leverage Contextual Insights\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(Sebastian and Richards \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) Sebastian \u0026amp; Richards, \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2017\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eQuasi-experimental pretest-posttest design\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eExamine whether education or contact interventions delivered by either ECAs or video presentations result in improvements in Mental Health Literacy postintervention compared to preintervention among the university students(N\u0026thinsp;=\u0026thinsp;245).\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1. Longitudinal Research\u003c/p\u003e \u003cp\u003e2. Clarify Stigmatization Process\u003c/p\u003e \u003cp\u003e3. Expand Intervention Scope\u003c/p\u003e \u003cp\u003e4. Enhance Agent Personalization\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(Vereschagin et al. \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) Vereschagin et al., \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2024\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA 2-arm, parallel-assignment, single-blinded, 30-day randomized controlled experimental research design\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eexamine the effectiveness of the\u0026nbsp;\u003cem\u003eMinder\u003c/em\u003e\u0026nbsp;mobile app in improving mental health and substance use outcomes in a general population of university students (N\u0026thinsp;=\u0026thinsp;1489)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1. Conduct Per-Protocol and Subgroup Analyses\u003c/p\u003e \u003cp\u003e2.Personalization of Content and Features\u003c/p\u003e \u003cp\u003e3. Enhance User Engagement\u003c/p\u003e \u003cp\u003e4. Implement Co-development and Participatory Design\u003c/p\u003e \u003cp\u003e5. System Integration\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(Trappey et al. \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) Trappey et al., \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2022\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eModule development and quasi experimental pretest-posttest design\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eDevelopment of an Empathy-Centric Counseling Chatbot System and Evaluate empathy-centric counseling in the chatbot system among the university student in Taiwan\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.Enhance the therapeutic realism of the system\u003c/p\u003e \u003cp\u003e2.Integrate more advanced language modeling\u003c/p\u003e \u003cp\u003e3.Expand intervention modules, and explore VRECC\u0026rsquo;s potential as a scalable virtual assistant for campus-based mental health services\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(Park et al. \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) Park et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2019\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eQualitative Case Study\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCreate a conversational sequence for a quick motivational interview to be delivered by a Web-based text messaging application (chatbot) and study its conversational experience with graduate students in their coping with stress.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1. Explore MI Sequences\u003c/p\u003e \u003cp\u003e2. Multi-Session Interventions\u003c/p\u003e \u003cp\u003e3. Larger Field Studies\u003c/p\u003e \u003cp\u003e4. Develop Assessment Toolkit\u003c/p\u003e \u003cp\u003e5. Incorporate Multimedia/Embodiment\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(Ehrlich et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) Ehrlich et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2024\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA two arm, randomized, controlled trial\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eEvaluate the Mind Tutor, an AI-enhanced tool created to boost student well-being.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1. Link Usage Patterns to Outcomes\u003c/p\u003e \u003cp\u003e2. Understand User Interaction\u003c/p\u003e \u003cp\u003e3. Tailor App Development\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(Gbollie et al. \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) Gbollie et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2023\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eExploratory factor analysis and multivariate ordinal regression\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eexamine university students (\u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;17 838)\u0026rsquo; experiences with, attitudes toward, and intentions to use digital mental health solutions, and to identify key factors influencing their intention to use these tools.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1. Investigate low engagement with chatbots and develop strategies to improve uptake\u003c/p\u003e \u003cp\u003e2. Develop psychoeducational interventions\u003c/p\u003e \u003cp\u003e3. Expand research samples and national representation\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(Abdaljaleel et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) Abdaljaleel et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2024\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003estructured survey\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCompare college students' opinions of counseling offered by human counselors to counseling generated by Pi, a helpful and sympathetic chatbot.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1. Identify Key Factors led to the positive perception and inability to distinguish AI from human\u003c/p\u003e \u003cp\u003e2. Complex \u0026amp; Extended Interactions\u003c/p\u003e \u003cp\u003e3. Investigate AI's Multimodal Communication (facial expressions, vocal tones)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eCluster 1: Chatbot use for academic stress management is still limited by technology. Future research must focus on advancing NLP, emotion recognition, and conversational flow. Personalizing interventions to individual user needs is also critical. Studies require larger, more diverse samples and longitudinal designs to assess sustained impact. Collaborating with universities to integrate chatbots into student support systems is recommended.\u003c/p\u003e \u003cp\u003eCluster 2: Future research should utilize clinically confirmed samples to address self-selection bias and clarify therapeutic boundaries for diagnosed populations. Direct comparisons with traditional therapy are essential to assess efficacy. Advancing multimodal interactions (text, voice, visuals) remain crucial for engagement. Strict adherence to ethical standards, particularly regarding privacy, informed consent, and safeguarding is imperative.\u003c/p\u003e \u003cp\u003eCluster 3: Future research must prioritize participatory design, actively involving both students and mental health professionals. Given demonstrated cross-cultural variations in effectiveness, cultural and contextual adaptation is essential. Ensuring acceptability and efficacy requires ongoing user feedback and broad-scale testing. Developing scalable, culturally sensitive systems is key for wider implementation.\u003c/p\u003e \u003cp\u003eCluster 4: Rigorous experimental designs including control groups, blinding, stratified randomization, and longitudinal follow-ups are fundamental for robust evaluation. Interventions should adopt multi-session, long-term approaches to measure sustained outcomes. Improved metrics for engagement, fidelity, and satisfaction are needed, alongside mixed methods approach. Integrating novel technologies (e.g., VR, motivational interviewing, hybrid human-AI) can enhance interventions. Developing modular systems targeting specific psychological goals (e.g., emotion regulation) is recommended.\u003c/p\u003e \u003cp\u003eCluster 5: Future research must incorporate sociocultural factors in design and evaluation, as user experience varies significantly across cultures. Despite this variation, enhancing engagement through technological innovation is consistently important. Understanding determinants of positive experience and mitigating factors for low retention is critical. Chatbots supporting sustained, context-aware interactions via multimodal communication, personalization, and privacy features enhance satisfaction and long-term adherence.\u003c/p\u003e \u003c/div\u003e"},{"header":"5. Discussion","content":"\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e5.1 Overall findings from bibliometric analysis\u003c/h2\u003e \u003cp\u003eThe evolution of publication volume over time closely reflects the technological advancements that underpin the development of AI-based chatbots. One of the earliest relevant studies, conducted by Kavakli et al. (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2012\u003c/span\u003e), emphasized the potential of human‒computer interaction (HCI) to enable \u0026ldquo;virtual humans\u0026rdquo; to act as virtual psychologists. However, at that time, the field of AI had not yet focused on modeling expressive agent motions capable of conveying personality and nonverbal communication effectively. A technological shift occurred approximately 2017\u0026ndash;2018, marked by improvements in voice recognition and conversational interfaces, which made conversational agents more accessible and functional (Fitzpatrick et al. 2017b). This period also resulted in a corresponding increase in related publications. A major turning point arrived at the end of 2022 with the rapid development of generative large language models (LLMs) (Khan et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), which significantly increased the capacity for personalized interaction. Many of the reviewed studies acknowledged this technological leap, engaged in natural, multiturn conversations that emulated human empathy, understanding, and responsiveness (Al-Amin et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). These advances, combined with the mental health challenges exacerbated by the COVID-19 pandemic (Aristovnik et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), may contributed to a sharp rise in publications after 2021, consistent with global reports of increased interest in digital mental health solutions during and after the pandemic (He et al. \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Marwaha et al. \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn terms of research geography, China, South Korea, and the United States lead in total publication volume. However, no institutions from these countries rank among the top three most productive institutions, suggesting relatively limited domestic and inter-institutional collaboration. In contrast, institutions from Canada, Australia, and Jordan are among the most productive, indicating robust national-level and interinstitutional collaboration. Current research remains concentrated in developed and upper-middle-income countries, with minimal representation from low-income or underdeveloped regions. This geographic imbalance mirrors the need for more inclusive and geographically diverse research to address disparities in access to technological health interventions (Hoagland and Kipping \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe prominence of \u0026ldquo;AI\u0026rdquo; and \u0026ldquo;ChatGPT\u0026rdquo; as the top keywords confirms that generative AI, particularly ChatGPT has become the most frequently studied chatbot in this field. Alongside these developments, mobile health (mHealth) applications have also attracted notable attention, indicating a broader trend toward technological integration in mental health care. High-frequency mental health-related keywords such as \u0026ldquo;anxiety\u0026rdquo; and \u0026ldquo;depression\u0026rdquo; align with findings from recent studies during or shortly after the COVID-19 pandemic, when the prevalence of these conditions reportedly rose by about 25% worldwide (WHO, 2022). The pandemic heightened concerns on student well-being and normalized mental health discourse within higher education (Harris et al. \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The recurring term \u0026ldquo;stress management\u0026rdquo; underscores the widespread presence of academic stress among students and the recognition of chatbots as potential tools for its alleviation (Mofatteh, \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Collectively, these keyword patterns emphasize the centrality of stress, anxiety, and depression in both developed and developing contexts, and highlight the importance of embedding digital technologies into campus-based prevention and intervention strategies (S. Zhai et al. \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). The frequent appearance of \u0026ldquo;intervention\u0026rdquo; further suggests that much of the literature is experimental or quasi-experimental in nature, reflecting the rapidly evolving state of AI-based mental health applications in higher education.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e5.2 Future directions related to SDGs based on the systematic literature review and beyond\u003c/h2\u003e \u003cp\u003eBased on systematic literature review and the preceding bibliometric analysis, the following future research directions integrate micro-level recommendations, focused on research design, technology, user engagement, and cultural adaptation, with macro-level strategies, such as global collaboration, institutional integration, and ethical governance. Framing these directions within the United Nations SDGs highlights their relevance not only to academic and technological advancement but also to sustainable global well-being.\u003c/p\u003e \u003cp\u003eFrom the perspectives of research design and intervention evaluation, many studies emphasize the need for more rigorous experimental designs. Research design shift strengthens the evidence base for effective mental health interventions in higher education, which aligns with SDG 3 (Good Health and Well-being) and SDG 4 (Quality Education). Randomized controlled trials (RCTs) are widely regarded as the most reliable methods for establishing causal relationships (Krauss \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Meanwhile, we should conduct more longitudinal studies to track long-term effects and psychological and educational research, where short-term or single-session interventions without follow-up often limit the generalizability and durability of the results. Furthermore, although some studies have incorporated user feedback through qualitative methods or postintervention surveys, the number of mixed-method studies remains limited. Expanding the use of mixed-method designs could enrich the evaluation of both the psychological impact and the user experience.\u003c/p\u003e \u003cp\u003eFrom a technological perspective, there is a need for technology improvement and make breakthroughs in generative AI models. This direction supports SDG 9 (Industry, Innovation, and Infrastructure) and SDG 3, fostering innovation that directly benefits mental health outcomes. As chatbots become more advanced, future studies should focus on enhancing natural language processing (NLP), emotion recognition, and multimodal interaction capabilities to make interventions more responsive, empathetic, and context aware (Babu et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). With respect to design-related concerns, a recurring issue is low user engagement, particularly in mental health apps, such as that mentioned by Ehrlich et al. (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), poor engagement not only limits the effectiveness of interventions but also hampers efforts to evaluate outcomes meaningfully. A frequently reported form of user complaint is the lack of personalization, as Fitzpatrick et al. (2017b) attempted to explore how to improve user engagement via personalization, and the latest studies recommended analyzing user input and behavioral data (Manole et al. \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) or tailor interaction (Ehrlich et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Some studies have proposed participatory or codesign approaches as potential solutions, allowing users to contribute to the design and refinement of chatbot functions, content, and delivery style (Grov\u0026eacute; \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Potts et al. 2021b). Additionally, the review indicates that future development of AI-based chatbots should consider the local sociocultural background of target populations, not only by translating chatbot content into local languages but also by aligning tone, values, and support strategies with cultural expectations and norms (C. Zhai and Wibowo \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). These strategies offer valuable insights into the future development of more adaptive and engaging mental health applications, these strategies advance SDG 10 (Reduced Inequalities), ensuring equitable and culturally relevant support.\u003c/p\u003e \u003cp\u003eFrom a macro perspective, it is essential to reduce the global divide in AI research and implementation. The World Health Organization (2020) has highlighted the potential of AI to bridge health gaps in low- and middle-income countries (LMICs). The current Bibliometric evidence indicates that research output is concentrated in a limited number of technologically advanced countries, creating a risk of widening global disparities in access to AI-based mental health tools. Such issues directly address SDG 17 (Partnerships for the Goals) and SDG 10 (Reduced Inequalities), promoting inclusive and cooperative innovation (Nations \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Alami et al. (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) further proposed training and retention of local expertise, a comprehensive monitoring system, a systems-based approach to implementation, and accountable local leadership inclusive of all stakeholders. Therefore, international partnerships through joint projects, shared datasets, and capacity-building programs is need (Mhlanga \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Beyond geographical inequities, health equity must also be extended to marginalized university student groups such as ethnic minorities, immigrants, people with disabilities, and sexual minorities. For example, Nadarzynski et al. (\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) noted that stigma and discomfort often hinder these populations from discussing sensitive issues, highlighting the importance of culturally sensitive chatbot design.\u003c/p\u003e \u003cp\u003eAnother critical future direction is strengthening the cooperation network between AI chatbot developers and universities. Many reviewed studies have emphasized the need for campus-specific applications that address the daily experiences and psychosocial challenges of students (Ehrlich et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Vereschagin et al. \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). For some mental health App developers should work closely with university counseling centers to facilitate ethically responsible data collection, crisis sample screening, and timely interventions for students seeking psychological support (Lattie et al. \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). In parallel, collaboration with university administrations is essential to promote awareness campaigns, ensure institutional endorsement, and foster the adoption of AI-based tools among students. Moreover, chatbot-related competencies could be integrated into mental health programs and curricula, equipping students with skills to effectively use digital tools for self-help and preventive care, students who have received sufficient mental health education are better prepared to make educated healthcare decisions in the future, stressing the long-term influence of educational interventions (S. Zhai et al. \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Finally, developers should leverage the research resources available within universities, including longitudinal student data and campus-based pilot studies, to enable evidence-driven refinement and monitoring of interventions. Such systemic integration would not only enhance the ecological validity of chatbot applications but also advance SDG 3 (Good Health and Well-being) by improving access to mental health care, and SDG 4 (Quality Education) by embedding innovative support systems into higher education environments.\u003c/p\u003e \u003cp\u003eEthical issues should be greatly concerned regarding deployment of AI in mental health. Many reviewed studies mentioned ethical safeguard, privacy, data security, and algorithmic bias are particularly pressing in the context of vulnerable populations such as university students (Fulmer et al. \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Olla et al. \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). To deal with this issue, it\u0026rsquo;s recommended to obey data privacy laws where necessary, or make sure that user know how their data is being used, whether that means just storing it or using it to train AI models, or making sure that patients know how their data is being stored and kept safe (Coghlan et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Moreover, the cases are more complex than we expected, some contended that users could be discouraged from obtaining appropriate mental health services if they receive inappropriate responses (Vilaza and McCashin \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), or users\u0026rsquo; dependence and liable to manipulation on the chatbots (Laestadius et al. \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Coghlan et al. (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) mentioned five AI ethics principles, (a) non-maleficence, (b) beneficence, (c) respect for autonomy, (d) justice and (e) explicability. These principles are quite relevant to the AI technology design and the protection from the law, which also need the supervision from the human (Coghlan et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), however, this topic is still underdiscussed and need to be further researched. Although some governments established some policies to monitor (Martinez-Martin \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), but AI Chabot still has a lot of room for improvement on some topics such as the boundaries of psychological interventions, the discussion of sensitive topics, and the standardization and rationalization of the accountability system (Vilaza and McCashin \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), and it is also necessary for policy makers and developers, psychologists, legal practitioners, university administrators, and university users to participate in the supplementation of the ethical framework of the AI-chatbot together. These priorities align with SDG 16 (Peace, Justice, and Strong Institutions), which underscores the importance of trustworthy governance, and with SDG 3, as ethical safeguards are integral to protecting student well-being. Building trust through ethical governance will not only enhance user acceptance but also ensure that AI chatbots contribute positively to mental health support in ways that are safe, equitable, and sustainable.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e5.3 Implications\u003c/h2\u003e \u003cp\u003eThis study advances theoretical understanding of digital mental health and AI-based chatbots within higher education context. By integrating bibliometric mapping with a systematic literature review, it bridges technological perspectives from computer science with psychological and educational frameworks. This study reveals emerging affective dimensions in human‒chatbot interactions, suggesting an expansion of traditional human‒computer interaction frameworks to incorporate digital empathy, trust-building, and emotional regulation. Importantly, the study situates these contributions within the Sustainable Development Goals (SDGs), particularly SDG 3 (Good Health and Well-being) and SDG 4 (Quality Education), SDG 10 (Reduced Inequalities), SDG 16 (Peace, Justice, and Strong Institutions) and SDG 17 (Partnerships for the Goals) reinforcing the role of AI-based mental health tools in supporting equitable access to psychological support for university students.\u003c/p\u003e \u003cp\u003eThe findings suggest that policymakers should consider integrating chatbots as a complement to traditional services into institutional mental health strategies (Olla et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). They should actively encourage the use of therapeutic chatbots as a complement to traditional campus services, providing professional guidance from mental health practitioners to help students identify high-quality resources and utilize them effectively for psychological well-being (Gao \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Oghenekaro and Okoro \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Also, they should establish regulatory frameworks that ensure privacy, data protection, and ethical accountability in AI-driven mental health tools. Promoting equitable funding mechanisms and capacity-building programs. Researchers should establish a research collaboration network and design rigorous experimental studies including randomized controlled trials (RCTs), longitudinal research, cross-disciplinary innovation and cultural adaptation research. For developers and technology providers, they should enhance NLP capabilities, multimodal interaction, and adaptive learning systems, engage in participatory co-design with students, educators, psychologists, counselors and lawyers. International organizations should facilitate global partnerships and knowledge exchange, supporting LMICs in adopting AI for mental health, they should also set ethical benchmarks and foster open-access research databases, ensuring global inclusivity.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e5.4 Limitations\u003c/h2\u003e \u003cp\u003eThis study is based on data retrieved from the Web of Science Core Collection and Scopus, which ensures a focus on high-quality publications. However, this also means that relevant studies indexed in other databases, such as PubMed, may have been excluded. As a result, the overall number of analyzed publications may be limited, potentially affecting the comprehensiveness of the findings.\u003c/p\u003e \u003cp\u003eFurthermore, our systematic review focused on highly cited papers from bibliometric analysis. This strategy may have excluded newer or lower-cited studies, meaning some emerging insights could be missed. Consequently, the future directions identified are drawn from influential works and should be interpreted in light of this selection bias.\u003c/p\u003e \u003cp\u003eAdditionally, the exclusion of non-English language articles means that studies published in other languages, such as Chinese or Spanish, were not included. This could lead to a regional bias in the interpretation of research distribution, particularly underrepresentation of contributions from countries such as China and those in Latin America.\u003c/p\u003e \u003cp\u003eFinally, the current study focused only on university students, which limits its generalizability to other populations, such as clinical samples or broadened age groups.\u003c/p\u003e \u003c/div\u003e"},{"header":"6. Conclusion","content":"\u003cp\u003eThis study combined bibliometric analysis and a systematic literature review to explore the landscape of AI-based chatbot research in supporting university students' mental health. The findings revealed a growing interest in this field, with emerging research clusters focusing on academic stress, emotional support, high-risk mental health scenarios, intervention design, and the user experience. Our findings highlight the link between digital mental health interventions and SDGs and practical implications for higher education. Researchers should pursue more longitudinal and cross-cultural studies, policymakers and administrators need to establish supportive regulations and institutional frameworks. Developers must ensure inclusive design and data transparency. Moving forward, global collaboration is indispensable. This emerging research area associates with key Sustainable Development Goals, and continued development has the potential to contribute to student well-being and broader SDG targets.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003eNo funding applicable\u003c/p\u003e \u003cp\u003eData Availability\u003c/p\u003e \u003cp\u003eAll data generated or analyzed during this study are included in this published article and its supplementary materials. Raw bibliometric data were retrieved from the Web of Science and Scopus database. Additional data are available from the corresponding author upon reasonable request.\u003c/p\u003e \u003cp\u003eCompeting Interests\u003c/p\u003e \u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e \u003cp\u003eEthical Statements\u003c/p\u003e \u003cp\u003eThis article does not contain any studies with human participants or animals performed by any of the authors. Ethical approval was therefore not required.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eAll authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Yi Xing, Chen Ying, Loh Sau Cheong and Siaw Yan Li. The first draft of the manuscript was written by Yi Xing and all authors commented on previous versions of the manuscript. The final version was proofread by Hao Li Jie and Hassan Abuhassna. All authors read and approved of the final manuscript.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eAll data generated or analyzed during this study are included in this published article and its supplementary materials. Raw bibliometric data were retrieved from the Web of Science and Scopus database. Additional data are available from the corresponding author upon reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAbdaljaleel M et al (2024) 'A multinational study on the factors influencing university students\u0026rsquo; attitudes and usage of ChatGPT', \u003cem\u003eScientific Reports\u003c/em\u003e, 14 (1), 1983. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-024-52549-8\u003c/span\u003e\u003cspan address=\"10.1038/s41598-024-52549-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbuhassna H et al (2024) The Information Age for Education via Artificial Intelligence and Machine Learning: A Bibliometric and Systematic Literature Analysis. Int J Inform Educ Technol 14:700\u0026ndash;711. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.18178/ijiet.2024.14.5.2095\u003c/span\u003e\u003cspan address=\"10.18178/ijiet.2024.14.5.2095\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAdamopoulou E, Moussiades L (2020) An Overview of Chatbot Technology. In: Maglogiannis I, Iliadis L, Pimenidis E (eds) Artificial Intelligence Applications and Innovations, 2 edn. Springer International Publishing), Greece, pp 373\u0026ndash;383. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/978-3-030-49186-4_31\u003c/span\u003e\u003cspan address=\"10.1007/978-3-030-49186-4_31\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003eNeos Marmaras\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAl-Amin M et al (2024) 'History of generative Artificial Intelligence (AI) chatbots: past, present, and future development', \u003cem\u003earXiv preprint arXiv:2402.05122\u003c/em\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlami H et al (2020) Artificial intelligence in health care: laying the Foundation for Responsible, sustainable, and inclusive innovation in low- and middle-income countries. Globalization Health 16(1):52. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12992-020-00584-1\u003c/span\u003e\u003cspan address=\"10.1186/s12992-020-00584-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAristovnik A et al (2020) Impacts of the COVID-19 pandemic on life of higher education students: A global perspective. Sustainability 12(20):8438. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/su12208438\u003c/span\u003e\u003cspan address=\"10.3390/su12208438\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAuerbach, Randy P et al (2018) WHO world mental health surveys international college student project: Prevalence and distribution of mental disorders. J Abnorm Psychol 127(7):623. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1037/abn0000362\u003c/span\u003e\u003cspan address=\"10.1037/abn0000362\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBabu A et al (2024) 'Multimodal Emotion Analysis Using Integrating NLP, AI, and Facial Expression Recognition for Enhanced Emotion Detection', \u003cem\u003e2024 IEEE International Conference on Signal Processing, Informatics, Communication and Energy Systems (SPICES)\u003c/em\u003e, 1\u0026ndash;6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1109/SPICES62143.2024.10779754\u003c/span\u003e\u003cspan address=\"10.1109/SPICES62143.2024.10779754\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBansal G et al (2024) Transforming conversations with AI\u0026mdash;A comprehensive study of ChatGPT. Cogn Comput 16(5):2487\u0026ndash;2510. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi/10.1007/s12559-023-10236-2\u003c/span\u003e\u003cspan address=\"https://doi/10.1007/s12559-023-10236-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChan CKY (2025) AI as the Therapist: Student Insights on the Challenges of Using Generative AI for School Mental Health Frameworks. Behav Sci (Basel) 15(3). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/bs15030287\u003c/span\u003e\u003cspan address=\"10.3390/bs15030287\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCoghlan S et al (2023) To chat or bot to chat: Ethical issues with using chatbots in mental health. Digit Health 9:20552076231183542. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/20552076231183542\u003c/span\u003e\u003cspan address=\"10.1177/20552076231183542\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDenecke K, Gabarron E (2024) The ethical aspects of integrating sentiment and emotion analysis in chatbots for depression intervention. Front Psychiatry 15:1462083. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fpsyt.2024.1462083\u003c/span\u003e\u003cspan address=\"10.3389/fpsyt.2024.1462083\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDonthu N et al (2021) How to conduct a bibliometric analysis: An overview and guidelines. J Bus Res 133:285\u0026ndash;296. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jbusres.2021.04.070\u003c/span\u003e\u003cspan address=\"10.1016/j.jbusres.2021.04.070\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEhrlich C et al (2024) Evaluation of an Artificial Intelligence Enhanced Application for Student Wellbeing: Pilot Randomised Trial of the Mind Tutor. Int J Appl Posit Psychol 9(1):435\u0026ndash;454. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s41042-023-00133-2\u003c/span\u003e\u003cspan address=\"10.1007/s41042-023-00133-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEltahawy L et al (2024) Can robots do therapy? Examining the efficacy of a CBT bot in comparison with other behavioral intervention technologies in alleviating mental health symptoms. Computers Hum Behavior: Artif Hum 2(1):100035. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.chbah.2023.100035\u003c/span\u003e\u003cspan address=\"10.1016/j.chbah.2023.100035\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFarzan M et al (2025) Artificial intelligence-powered cognitive behavioral therapy chatbots, a systematic review. Iran J psychiatry 20(1):102. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.18502/ijps.v20i1.17395\u003c/span\u003e\u003cspan address=\"10.18502/ijps.v20i1.17395\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFeng Y et al (2025) Effectiveness of AI-Driven Conversational Agents in Improving Mental Health Among Young People: Systematic Review and Meta-Analysis. J Med Internet Res 27:e69639. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2196/69639\u003c/span\u003e\u003cspan address=\"10.2196/69639\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFitzpatrick KK, Darcy, Alison, Vierhile M (2017a) Delivering cognitive behavior therapy to young adults with symptoms of depression and anxiety using a fully automated conversational agent (Woebot): a randomized controlled trial. JMIR mental health 4(2):e7785. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2196/mental.7785\u003c/span\u003e\u003cspan address=\"10.2196/mental.7785\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFulmer R et al (2018) Using Psychological Artificial Intelligence (Tess) to Relieve Symptoms of Depression and Anxiety: Randomized Controlled Trial. JMIR Ment Health 5(4):e64. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2196/mental.9782\u003c/span\u003e\u003cspan address=\"10.2196/mental.9782\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGabrielli S et al (2021) Engagement and Effectiveness of a Healthy-Coping Intervention via Chatbot for University Students During the COVID-19 Pandemic: Mixed Methods Proof-of-Concept Study. JMIR Mhealth Uhealth 9(5):e27965. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2196/27965\u003c/span\u003e\u003cspan address=\"10.2196/27965\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGao Y (2025) The impact and application of artificial intelligence technology on mental health counseling services for college students. J Comput Methods Sci Eng 25(2):1686\u0026ndash;1701. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/14727978241302641\u003c/span\u003e\u003cspan address=\"10.1177/14727978241302641\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGbollie E, Fayiah et al (2023) Intention to use digital mental health solutions: A cross-sectional survey of university students attitudes and perceptions toward online therapy, mental health apps, and chatbots. Digit HEALTH 9:20552076231216559. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/20552076231216559\u003c/span\u003e\u003cspan address=\"10.1177/20552076231216559\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrov\u0026eacute; C (2021) Co-developing a Mental Health and Wellbeing Chatbot With and for Young People. Front Psychiatry 11\u0026ndash;2020. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fpsyt.2020.606041\u003c/span\u003e\u003cspan address=\"10.3389/fpsyt.2020.606041\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHarris BR, Maher BM, Wentworth L (2022) Optimizing Efforts to Promote Mental Health on College and University Campuses: Recommendations to Facilitate Usage of Services, Resources, and Supports. J Behav Health Serv Res 49(2):252\u0026ndash;258. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11414-021-09780-2\u003c/span\u003e\u003cspan address=\"10.1007/s11414-021-09780-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHe Y et al (2022) Mental health chatbot for young adults with depressive symptoms during the COVID-19 pandemic: single-blind, three-arm randomized controlled trial. J Med Internet Res 24(11):e40719. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2196/40719\u003c/span\u003e\u003cspan address=\"10.2196/40719\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoagland A, Kipping S (2024) Challenges in promoting health equity and reducing disparities in access across new and established technologies. Can J Cardiol 40(6):1154\u0026ndash;1167. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.cjca.2024.02.014\u003c/span\u003e\u003cspan address=\"10.1016/j.cjca.2024.02.014\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHopman K, Richards D, Norberg MM (2023) A Digital Coach to Promote Emotion Regulation Skills. Multimodal Technol Interact 7(6):57. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/mti7060057\u003c/span\u003e\u003cspan address=\"10.3390/mti7060057\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHyseni Duraku, Zamira, Davis H, Hamiti E (2023) Mental health, study skills, social support, and barriers to seeking psychological help among university students: a call for mental health support in higher education. Front Public Health 11:1220614. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fpubh.2023.1220614\u003c/span\u003e\u003cspan address=\"10.3389/fpubh.2023.1220614\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJones WH (1985) 'The psychology of loneliness: Some personality issues in the study of social support', \u003cem\u003eSocial support: Theory, research and applications\u003c/em\u003e (Springer), 225\u0026thinsp;\u0026ndash;\u0026thinsp;41. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/978-94-009-5115-0_13\u003c/span\u003e\u003cspan address=\"10.1007/978-94-009-5115-0_13\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKaryotaki E et al (2020) Sources of stress and their associations with mental disorders among college students: results of the world health organization world mental health surveys international college student initiative. Front Psychol 11:1759. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fpsyg.2020.01759\u003c/span\u003e\u003cspan address=\"10.3389/fpsyg.2020.01759\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKavakli M, Li M, Rudra T (2012) Towards the Development of A Virtual Counselor to Tackle Students' Exam Stress. J Integr Des Process Sci 16(1):5\u0026ndash;26. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3233/jid-2012-0005\u003c/span\u003e\u003cspan address=\"10.3233/jid-2012-0005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKhan N et al (2024) Global insights and the impact of generative AI-ChatGPT on multidisciplinary: a systematic review and bibliometric analysis. Connection Sci 36(1):2353630. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/09540091.2024.2353630\u003c/span\u003e\u003cspan address=\"10.1080/09540091.2024.2353630\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKlos M, Carolina et al (2021) Artificial Intelligence\u0026ndash;Based Chatbot for Anxiety and Depression in University Students: Pilot Randomized Controlled Trial. JMIR Form Res 5(8):e20678. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2196/20678\u003c/span\u003e\u003cspan address=\"10.2196/20678\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKrauss A (2021) Assessing the Overall Validity of Randomised Controlled Trials. Int Stud Philos Sci 34(3):159\u0026ndash;182. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/02698595.2021.2002676\u003c/span\u003e\u003cspan address=\"10.1080/02698595.2021.2002676\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKroenke K, Spitzer RL, Williams, Janet BW (2001) The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med 16(9) 606\u0026thinsp;\u0026ndash;\u0026thinsp;13. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1046/j.1525-1497.2001.016009606.x\u003c/span\u003e\u003cspan address=\"10.1046/j.1525-1497.2001.016009606.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLaestadius L et al (2022) Too human and not human enough: A grounded theory analysis of mental health harms from emotional dependence on the social chatbot Replika. New Media Soc 26(10):5923\u0026ndash;5941. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/14614448221142007\u003c/span\u003e\u003cspan address=\"10.1177/14614448221142007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLattie EG et al (2022) Uptake and effectiveness of a self-guided mobile app platform for college student mental health. Internet interventions 27:100493. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.invent.2021.100493\u003c/span\u003e\u003cspan address=\"10.1016/j.invent.2021.100493\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi H et al (2023) Systematic review and meta-analysis of AI-based conversational agents for promoting mental health and well-being. NPJ Digit Med 6(1):236. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41746-023-00979-5\u003c/span\u003e\u003cspan address=\"10.1038/s41746-023-00979-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi Y et al (2025) Chatbot-Based Mindfulness-Based Stress Reduction Program for University Students With Depressive Symptoms: Intervention Development and Pilot Evaluation. J Am Psychiatr Nurses Assoc 31(4):398\u0026ndash;411. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/10783903241302092\u003c/span\u003e\u003cspan address=\"10.1177/10783903241302092\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu H et al (2022) Using AI chatbots to provide self-help depression interventions for university students: A randomized trial of effectiveness. Internet Interv 27:100495. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.invent.2022.100495\u003c/span\u003e\u003cspan address=\"10.1016/j.invent.2022.100495\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu Z et al (2015) Visualizing the intellectual structure and evolution of innovation systems research: a bibliometric analysis. Scientometrics 103:135\u0026ndash;158. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11192-014-1517-y\u003c/span\u003e\u003cspan address=\"10.1007/s11192-014-1517-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaciejewski J, Smoktunowicz E (2023) Low-effort internet intervention to reduce students' stress delivered with Meta's Messenger chatbot (Stressbot): A randomized controlled trial. Internet Interventions 33:100653. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.invent.2023.100653\u003c/span\u003e\u003cspan address=\"10.1016/j.invent.2023.100653\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eManole A et al (2024) 'Harnessing AI in Anxiety Management: A Chatbot-Based Intervention for Personalized Mental Health Support', \u003cem\u003eInformation\u003c/em\u003e, 15 (12)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMartinez-Martin N (2023) Viewing CAI as a Tool Within the Mental Health Care System. Am J Bioeth 23(5):57\u0026ndash;59. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/15265161.2023.2191058\u003c/span\u003e\u003cspan address=\"10.1080/15265161.2023.2191058\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarwaha EB et al (2025) The Digital Bridge: using Digital Intervention to Understand Psychosocial Correlation between COVID-19 Stress and Sleep Quality in Undergraduate Students Before and After COVID-19. Open Psychol J 18. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2174/0118743501364965250220093932\u003c/span\u003e\u003cspan address=\"10.2174/0118743501364965250220093932\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarzi G et al (2025) Guidelines for Bibliometric-Systematic Literature Reviews: 10 steps to combine analysis, synthesis and theory development. Int J Manage Reviews 27(1):81\u0026ndash;103. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/ijmr.12357\u003c/span\u003e\u003cspan address=\"10.1111/ijmr.12357\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMhlanga D (2023) 'The power of smart technologies for global partnership for development', \u003cem\u003eFinTech and artificial intelligence for sustainable development: The role of smart technologies in achieving development goals\u003c/em\u003e (Springer), 291\u0026ndash;315. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/978-3-031-37776-1_13\u003c/span\u003e\u003cspan address=\"10.1007/978-3-031-37776-1_13\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMofatteh M (2021) Risk factors associated with stress, anxiety, and depression among university undergraduate students. AIMS Public Health 8(1):36\u0026ndash;65. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3934/publichealth.2021004\u003c/span\u003e\u003cspan address=\"10.3934/publichealth.2021004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoldt JA et al (2022) Assessing medical students' perceived stress levels by comparing a chatbot-based approach to the Perceived Stress Questionnaire (PSQ20) in a mixed-methods study. Digit Health 8:20552076221139092\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNadarzynski T et al (2024) Achieving health equity through conversational AI: A roadmap for design and implementation of inclusive chatbots in healthcare. PLOS Digit Health 3(5):e0000492. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pdig.0000492\u003c/span\u003e\u003cspan address=\"10.1371/journal.pdig.0000492\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNations U (2015) 'Transforming our world: The 2030 Agenda for Sustainable Development'. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://sdgs.un.org/2030agenda\u003c/span\u003e\u003cspan address=\"https://sdgs.un.org/2030agenda\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNelekar S et al (2022) Effectiveness of embodied conversational agents for managing academic stress at an Indian University (ARU) during COVID-19. Br J Edu Technol 53(3):491\u0026ndash;511. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/bjet.13174\u003c/span\u003e\u003cspan address=\"10.1111/bjet.13174\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOghenekaro LU, Okoro CO (2024) Artificial intelligence-based chatbot for student mental health support. Open Access Libr J 11(5):1\u0026ndash;14. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.4236/oalib.1111511\u003c/span\u003e\u003cspan address=\"10.4236/oalib.1111511\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOlla P et al (2025) Deploying a Mental Health Chatbot in Higher Education: The Development and Evaluation of Luna, an AI-Based Mental Health Support System. Computers 14(6):227. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/computers14060227\u003c/span\u003e\u003cspan address=\"10.3390/computers14060227\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOmarov B, Narynov S, Zhumanov Z (2023) Artificial Intelligence-Enabled Chatbots in Mental Health: A Systematic Review. Computers Mater Continua 74(3):5105\u0026ndash;5122. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.32604/cmc.2023.034655\u003c/span\u003e\u003cspan address=\"10.32604/cmc.2023.034655\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePark SH et al (2019) Designing a Chatbot for a Brief Motivational Interview on Stress Management: Qualitative Case Study. J Med Internet Res 21(4):e12231. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2196/12231\u003c/span\u003e\u003cspan address=\"10.2196/12231\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePotts C et al (2021a) Correction to: Chatbots to Support Mental Wellbeing of People Living in Rural Areas: Can User Groups Contribute to Co-design? J Technol Behav Sci 6(4):666\u0026ndash;666. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s41347-021-00222-6\u003c/span\u003e\u003cspan address=\"10.1007/s41347-021-00222-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePriestley M et al (2022) Student perspectives on improving mental health support services at university. Counselling Psychother Res 22(1). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/capr.12391\u003c/span\u003e\u003cspan address=\"10.1002/capr.12391\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSalamanca-Sanabria A et al (2023) Exploring the Perceptions of mHealth Interventions for the Prevention of Common Mental Disorders in University Students in Singapore: Qualitative Study. J Med Internet Res 25:e44542. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2196/44542\u003c/span\u003e\u003cspan address=\"10.2196/44542\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSebastian J, Richards D (2017) Changing stigmatizing attitudes to mental health via education and contact with embodied conversational agents. Comput Hum Behav 73:479\u0026ndash;488. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.chb.2017.03.071\u003c/span\u003e\u003cspan address=\"10.1016/j.chb.2017.03.071\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSyed SA (2024) The role of AI in alleviating loneliness among adults in the United States. Int J Eng Technol Res Manage (IJETRM) 8(04):404\u0026ndash;421\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTrappey AJC et al (2022) 'Development of an Empathy-Centric Counseling Chatbot System Capable of Sentimental Dialogue Analysis', \u003cem\u003eProcesses\u003c/em\u003e, 10 (5). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/pr10050930\u003c/span\u003e\u003cspan address=\"10.3390/pr10050930\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVaidyam A, Nrusimha et al (2019) Chatbots and conversational agents in mental health: a review of the psychiatric landscape. Can J Psychiatry 64(7):456\u0026ndash;464. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/0706743719828977\u003c/span\u003e\u003cspan address=\"10.1177/0706743719828977\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVereschagin M et al (2024) Effectiveness of the Minder Mobile Mental Health and Substance Use Intervention for University Students: Randomized Controlled Trial. J Med Internet Res 26:e54287. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2196/54287\u003c/span\u003e\u003cspan address=\"10.2196/54287\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVilaza GN, McCashin D (2021) Is the Automation of Digital Mental Health Ethical? Applying an Ethical Framework to Chatbots for Cognitive Behaviour Therapy. Front Digit Health 3\u0026ndash;2021. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fdgth.2021.689736\u003c/span\u003e\u003cspan address=\"10.3389/fdgth.2021.689736\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhai C, Wibowo S (2022) 'A systematic review on cross-culture, humor and empathy dimensions in conversational chatbots: the case of second language acquisition', \u003cem\u003eHeliyon\u003c/em\u003e, 8 (12).\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.heliyon.2022.e12056\u003c/span\u003e\u003cspan address=\"10.1016/j.heliyon.2022.e12056\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhai S et al (2025) Technology-driven support: exploring the impact of artificial intelligence on mental health in higher education. Educ Inform Technol 30(10):13931\u0026ndash;13959. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10639-025-13352-8\u003c/span\u003e\u003cspan address=\"10.1007/s10639-025-13352-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou X et al (2021) Are online mental health interventions for youth effective? A systematic review. J Telemed Telecare 27(10):638\u0026ndash;666. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/1357633X211047285\u003c/span\u003e\u003cspan address=\"10.1177/1357633X211047285\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"humanities-and-social-sciences-communications","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"palcomms","sideBox":"Learn more about [Humanities \u0026 Social Sciences Communications](http://www.nature.com/palcomms/)","snPcode":"41599","submissionUrl":"https://submission.springernature.com/new-submission/41599/3","title":"Humanities and Social Sciences Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7486943/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7486943/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eUniversity students worldwide are experiencing increasing mental health challenges, while existing support resources remain insufficient. Artificial intelligence\u0026ndash;based chatbots have emerged as scalable and accessible tools for psychological support, aligning with the Sustainable Development Goals on health, quality education, and well-being. This study aims to map global research trends on AI-based chatbots for mental health in higher education and to identify future research directions. A bibliometric analysis was conducted using 58 English-language articles retrieved from Web of Science and Scopus. Publication patterns, influential journals, leading countries, and keyword networks were analyzed. Additionally, a systematic literature review of 20 selected studies was performed to synthesize research gaps and thematic priorities. The bibliometric analysis revealed a sharp increase in publications since 2020, with research concentrated in a limited number of countries and journals. Thematic synthesis highlighted five priority areas: stress management, mental health symptoms, emotional support, intervention design, and user interaction. However, the review also identified limited diversity in research designs and narrow applications of chatbot technologies. Findings demonstrate growing scholarly interest in AI-based chatbots for student well-being, yet significant gaps remain. Future research should focus on inclusive and user-centered chatbot design, integration of evidence-based interventions, international cooperation and cross-cultural validation, and ethical issues. These insights advance both theoretical understanding and practical development of AI-based mental health tools, contributing to the achievement of relevant Sustainable Development Goals.\u003c/p\u003e","manuscriptTitle":"AI Chatbots for University Student Mental Health: Bibliometric Mapping and Systematic Review with Future Directions for Sustainable Development Goals","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-16 08:25:18","doi":"10.21203/rs.3.rs-7486943/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-03-07T12:30:30+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-17T15:53:04+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-13T06:56:45+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"13738844297783489650057677301296637680","date":"2026-02-13T04:25:05+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"189156347663067625879550479624446329847","date":"2026-02-12T06:35:57+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-11T17:37:49+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"230113134442290680359031028682759315956","date":"2026-02-11T09:02:44+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"78152446183919502522747131918647669824","date":"2026-02-10T22:35:25+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-02-10T21:07:39+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-09T11:58:29+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-09-27T11:00:16+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-20T04:03:04+00:00","index":"","fulltext":""},{"type":"submitted","content":"Humanities and Social Sciences Communications","date":"2025-09-20T03:59:21+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"humanities-and-social-sciences-communications","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"palcomms","sideBox":"Learn more about [Humanities \u0026 Social Sciences Communications](http://www.nature.com/palcomms/)","snPcode":"41599","submissionUrl":"https://submission.springernature.com/new-submission/41599/3","title":"Humanities and Social Sciences Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"d6f5a355-e298-4d8c-8dad-59ef674d1457","owner":[],"postedDate":"February 16th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":62915205,"name":"Biological sciences/Psychology"},{"id":62915206,"name":"Social science/Psychology"},{"id":62915207,"name":"Scientific community and society/Scientific community"}],"tags":[],"updatedAt":"2026-05-15T09:15:56+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-16 08:25:18","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7486943","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7486943","identity":"rs-7486943","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}