A Machine Learning Model for the Prediction of Sexually Transmitted Diseases among the Youths in Southwestern Nigeria

A Machine Learning Model for the Prediction of Sexually Transmitted Diseases among the Youths in Southwestern Nigeria | 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 Research Article A Machine Learning Model for the Prediction of Sexually Transmitted Diseases among the Youths in Southwestern Nigeria Ozichi N. Emuoyibofarhe, Olubayode. Bamidele, Christian O. Osueke, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5404906/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Sexually transmitted diseases (STDs) are diseases which are spread between individuals through unprotected sexual contact. The spread has become rampant, especially among the youths nowadays who display promiscuous characteristics, which leads to a faster rate of the spread of the disease among the youths. Thus, this study aims to develop a machine learning model for an accurate analysis and prediction of the transmission rate of STDs among the youth within the southwestern region of Nigeria. For an approximate and optimize study, a questionnaire in Google form was administered to harvest opinions of youths within the stated demographic with respect to their health status, disease awareness, lifestyle choices and other characteristics. The collected primary dataset of 529 individual responses was used to build the machine learning model. The dataset was converted to comma-separated values (CSV) format to be trained and tested for a well-supervised machine learning model. Of the data collected, 75% served as training data and 25% served as testing data. Feature extraction, data visualization and data preprocessing were done to convert raw data into suitable machine learning. Taking from the results, a decision tree accuracy of 0.9393 with an area under the curve (AUC) score of 0.5843, logistic regression accuracy of 0.9621 with AUC score of 0.5, support vector machine accuracy of 0.96212 with AUC score of 0.5 and Gaussian Naïve Bayers machine learning algorithms accuracy score of 0.5909 with AUC score of 0.7874 were obtained. Hence, the Gaussian Naïve Bayers gave the best outcomes with an area under the curve (AUC) score of 0.79 and was able to correctly classify all 5 cases of STDs within the test set as compared to other algorithms. Machine learning model Sexually transmitted diseases Gaussian Naïve Bayers Logistic regression Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 INTRODUCTION Sexually transmitted diseases (STDs) are infections spread from infected individuals to other through unprotected sex. It is either a curable or incurable disease or infection that is primarily transmitted through unprotected sexual contact. Syphilis, gonorrhea, and Chlamydia Trachomatis are curable sexually transmitted infections (STIs), while herpes simplex, hepatitis B, human papillomavirus (HPV), and human immune virus (HIV) are incurable yet modifiable STIs, Azizi et al. [ 1 ]. Sexually transmitted diseases have been a persistent issue in Nigeria for as long as anybody can remember, and they do not appear to be going away anytime soon. In recent times, several sexually transmitted diseases (STDs) have been globally spread, which are majorly asymptomatic, Centers for Disease and Prevention [ 2 ]. Although there are no official statistics available, previous research has indicated that the prevalence of treatable STIs in Nigeria ranges from 0–18%, Daniel [ 3 ]. Though Western education is widely available in Nigeria's southwest states, the youths are either ignorant of the risks associated with unprotected sexual activity or prefer to ignore them. Additionally, research has indicated that youth often have a stronger propensity to engage in risky sexual behaviour with several partners due to their desire to explore their sexualities. As such, they are likely to have infectious diseases due to their careless attitude. STDs, the bulk of which are asymptomatic, are contracted daily around the world, ] National bureau of statistics [ 4 ]. Nigeria, being a highly populated black race, has about 200 million individuals with about 2.61% yearly birth rate increment, of which 16.8% are aged 15 to 35 grouped as youths, Centers for Disease and Prevention [ 2 ]. According to the Federal Ministry of Youth Development and the National Bureau of Statistics, the age group 15 to 35 years was appraised to be over 64 million individuals in 2012 (the most recent census). In the same census, it was estimated that the number of youths in the southwestern states of Nigeria was 13 million individuals, accounting for approximately 20% of the total population of youths in Nigeria, Daniel [ 3 ]. Therefore, the southwest states in Nigeria are Osun, Oyo, Ekiti, Ondo, and Ogun, which are the focus of this study. Machine learning approaches are widely used in the analysis and prediction of many different facets of human behaviour, Sri Preethaa et al [ 5 ]. The scientific investigation of mathematical theories and techniques applied to computing devices to carry out specified tasks without utilizing explicit instructions while concentrating on patterns and inferences is known as machine learning (ML), Emuoyibofarhe et al. [ 6 ]. Machine learning algorithms are employed to forecast the risk of sexual transmission among youths; and carry out certain task, which focuses on patterns and inferences rather than using explicit instructions, Ogunleye et al. [ 7 ]. A machine learning technique for the analysis of data computerizes the formulation of models. Artificial intelligence is created to make machines learn from data, spot patterns, and conclude with little human effort, Emuoyibofarhe et al. [ 8 ]. To create a machine learning algorithm, output predicted functions must be identified. As a result, this research aims to predict the sexual infectious disease rate among youths in southwestern Nigeria using a machine learning model, with the scope of the study limited to the prediction of STDs among adolescents. A machine learning tool-based model is developed to predict the sexual infections and HIV risk for one year. Utilizing models such as random forests, regression methods, support vector machines, and bagging ensembles, the tool analyzed data collected from March 2, 2015, to December 31, 2019. The final analysis reported the following incidence rates per 100 person-years: 16.95 [95% CI: 16.82–17.67] for gonorrhea, 17.96 [95% CI: 16.79–18.13] for chlamydia, and 0.21 [95% CI: 0.17–0.27] for syphilis and HIV, Adeboye et al. [ 9 ]. In a separate study, referenced in Xu et al. [ 10 ], machine learning techniques were employed to forecast testing HIV/STI and clinic attendance in response to clinical reminder messages. The study utilized data from 3,044 consultations to develop and refine the predictive models. A range of algorithms was implemented, including feedforward multi-layer neural artificial networks, generalized Bayesian linear models K-nearest neighbours, linear vector support machines, Naive Bayes, polynomial basis kernel, extreme gradient boosting, random forest, gradient boosting machines, and elastic net, ridge The median age of participants was 31.0 years, with reminder messages distributed via email (49.6%), SMS (30.6%), and other methods (19.8%). The study found that 15.5% of participants lived with HIV, and 19.1% had the symptoms of STI at the time of their clinic visit. Notably, 29.5% of patients attended the clinic within 30 days of the reminder. In Azizi et al. [ 1 ], the researchers developed a predictive risk-based model to analyze the impact of condoms on the sexual disease transmission rate. The results reveal that for a random mixing, the rising use of condoms will effectively decline the disease spread pattern in the population. Hence the model computed the risk of infection increase for individuals with many sexual partners, and the need for people with many sexual partners to use condoms for infection risk reduction. Nwadike et al. [ 11 ] undertook a study where records of patients diagnosed with sexually transmitted diseases were analyzed to recognize patterns of contraction of the infections within the patients. The records of 506 patients were used where females are 56.3% (285) and males are 43.7% (221) and their age ranges from one to eighty. In the analysis, the 1–10 age group and 71–80 age group seemed to be the lowest age group representation when the various features such as education level, sex, age, yeast cells presence, and disease venereal laboratory research test were used during the test. Furthermore, in Nzoputam et al. [ 12 ], the researchers stipulated the feature model essential chart that shows vagina itching and vagina discharge as the highest impact level on the patient diagnosed possibility with STDs, where 100% logistic regression shows a properly predicted model all the 101 true positives and 309 true negatives to have been wrongly diagnosed to be zero. According to Kiran et al. [ 13 ], Sepsis infection as the name implies is a risky infection initiated by the individual’s rejoinder to a disease, which normally brings about organ failure, damage of tissue, or death due to inflammation spreading all over the body. The results obtained by the researcher show that the machine learning algorithms give accurate prediction of sepsis with high specificity and sensitivity thereby giving an auspicious solution for primary clinical detection of sepsis. This study proposed a possible machine learning model for enhancing sepsis control and detection providing a basis for future study. Also, a study in Elder et al. [ 14 ] was carried out using a machine learning model to predict the risk of sexual infection among the youths in Southwestern Nigeria; in the study, data from a cross-sectional inspection of 2,000 people with age ranges from 15 to 24years were utilized to explore STD risk factors. Various machine learning algorithms, including vector support machines, decision trees, and logistic regression were employed for data analysis and to develop an analytical model. The resulting model, which incorporated features such as age, gender, sexual behaviour, and socioeconomic status, demonstrated high accuracy in predicting STD risk factors. The findings indicate that machine learning techniques hold significant potential for early detection and prevention of STDs among youth particularly in the resource-limited settings. A study conducted in Ethiopia applied machine learning techniques to identify predictors of sexually transmitted infections (STIs) and analyzed their geographic distribution across various regions. By understanding these predictors and their spatial patterns, policymakers can gain deeper insights into STI issues and tailor interventions more effectively. This research highlights the potential of machine learning in forecasting STIs within Ethiopia, Kassaw et al. [ 15 ]. Also, the study in Latt et al. [ 16 ] examined the potential of electronic health records for data routine to predict patients with a newly diagnosed sexually transmitted infection (STI) are most likely to contract another STI within the following 1 to 2 years. Furthermore, the study in Barman et al. [ 17 ] analyzed the disparity in the HIV distribution and sexually spread disease among various groups of populaces in Australia. Using Gini coefficients, the research found significant variations in HIV/STI risk, with higher Gini scores approaching one indicating greater inequality. Advancements in technology and the widespread use of the Internet, particularly among young people, have created valuable opportunities for leveraging eHealth in disease prevention. Nourimand et al. [ 18 ]. eHealth approaches can play a significant role in preventing sexually transmitted diseases (STDs), especially among the young demographic who frequently engage with new technologies and are at increased risk for STDs. Our systematic review specifically explored the preventative potential of eHealth for sexually transmitted infections (STIs), highlighting its various benefits and applications, WHO [ 19 – 21 ]. Also, a study stipulated that if the poor attitude or perception towards complying with the preventive measures continues, COVID-19 cases in Africa can result in an increase beyond the current spread. Adeyinka et al. [ 22 ] RESEARCH METHODOLOGY The research methodology adopted for the machine learning model for the prediction of sexually transmitted diseases among the youths in Southwestern Nigeria are respectively given. 2.1 Model Development A total of six machine learning models were executed on the dataset and the most appropriate model is chosen based on the assessment process. The details of how the entire process was done is detailed below: 2.1.1 Data Collection Primary data was the main and only source of data for the machine learning process. The data collection was done using the questionnaire data collection process via Google forms, which is a useful platform adopted to create and share digital survey for the research work. The platform had a function to convert the responses for easy analysis of the CSV format. A total of 529 responses were used to create the dataset for this study. A preview of the un-cleaned sample dataset from the google form is shown in Fig. 1 . 2.1.2 Data Loading and Analysis Feature Extraction The column containing the type of STD contracted, if the respondent has ever had an STD, is excluded because the dataset's output label is the STD column, and the specific type of STD is not considered a relevant feature for predicting the output label. Hence, the process of extracting this feature from the dataset is as shown below. Data Preprocessing During this stage, data cleaning was conducted, and string attributes were encoded to make them compatible with the algorithm. Microsoft excel facilitated the identification and replacement of these attributes. Subsequently, all string entities underwent label encoding, converting them into categorical numbers suitable for algorithmic processing. The preprocessing process is shown in the figure below: Above in Fig. 2 , all features were encoded. The outcome of this process reveals the newly cleaned dataset presented below. Implementation and Evaluation of Classification Algorithms The sanitized dataset is divided into two training dataset sections, 75% of the data serves as the training set, while the remaining 25% serves as the testing set. Columns one through seventeenth are designated as input labels, whereas the eighteenth column, represents the STD status which is designated as the output label. The training and testing data were utilized across all learning algorithms in subsequent processes using the Support Vector Classification (SVC) method. The implementation of the SVC algorithm yielded a model with an accuracy of 0.96212 and an AUC score of 0.5. The corresponding confusion matrix is also carried out. The confusion matrix in Fig. 3 indicates that the algorithm successfully classified all instances where individuals did not have the disease but failed to correctly identify any cases where individuals did have the disease. The AUC score of 0.5 further demonstrates the algorithm's poor performance, making it unsuitable for this classification problem. 3.0. Decision Tree Classification This is a method that is used to diagnose Sexually Transmitted Diseases (STDs) by analyzing the patient data such as the symptoms and test results. This is done by actually splitting the data at each node which is based on the availability of the specific features thereby leading to a final classification at the leaf nodes which in turn helps to identify the particular STD or stating that there is no infection. The method is straightforward, easy to interpret and highlights and also provides a quick and effective way to assist healthcare providers in the diagnostics of STDs. Model Evaluation The implemented algorithm achieved a correctness of 0.9393, an AUC score of 0.5843, an exactness of 0.9685, and a recall of 0.9685. The confusion matrix for this model is examined as presented in Fig. 4 . The confusion matrix revealed that the algorithm successfully classified 123 negative STD cases but correctly identified only one out of five positive STD cases. With an AUC of 0.5843, the model demonstrates limited performance. The ROC curve for this algorithm is illustrated in Fig. 5 below. 4.0 Gaussian Naïve Bayers Algorithm This method is particularly useful as it can incorporate local data on STD prevalence and common symptoms and is used to diagnose Sexually Transmitted Diseases (STDs) in this work by leveraging patient data with the underlined patient features based on the symptoms. Hence, the model attained an accuracy of 0.5909, a correctness of 1.0, a recall of 0.5748, and an AUC score of 0.7874. The AUC score of 0.7874 indicates that the algorithm is a stronger classifier than other algorithms when compared. The misperception matrix of the algorithm is shown in Fig. 6 below and the ROC curve for the algorithm is illustrated in Fig. 7 . The confusion matrix for the algorithm classifying the five types of STDs can be represented using the analysis below: Actual/Predicted STD 1 STD 2 STD 3 STD 4 STD 1 C11 C12 C13 C14 C15 STD 2 C21 C22 C23 C24 C25 STD 3 C31 C32 C33 C34 C35 STD 4 C41 C42 C43 C44 C45 STD 5 C51 C52 C53 C54 C55 Interpretation Diagonal Elements (Cii​): These represent the correctly classified instances for each class. For example, C11​ is the number of correctly predicted cases of STD 1. Off-Diagonal Elements (Cij where I ≠ j): These represent the misclassified instances. For instance, C12 is the number of STD 1 cases that were incorrectly predicted as STD 2. Performance Analysis From the confusion matrix, we can derive various performance metrics to assess the classification ability of the algorithm. Common metrics include: Accuracy : The overall proportion of correctly classified instances. Accuracy = \(\:\frac{\sum\:_{i=1}^{N}C11}{\sum\:_{i=1\:}^{N}\sum\:_{j=1}^{N}Cij}\) Precision for Class k: The correct proportion of the predicted occurrences of class k out of all occurrences foreseen as class k. Precision = ​​ \(\:\frac{{C}_{KK}}{\sum\:_{j=1}^{N}{C}_{ij}}\) Recall for Class k: The correct quantity of the forecast instances of class k out of all definite instances of class k. Recall = \(\:\frac{{C}_{kk}}{\sum\:_{J=1}^{N}{C}_{Kj}}\) F1-Score for class k: The harmonic means of exactness and recall for class k. F 1 – Score k = \(\:\frac{{Precision}_{k}-\:{Recall}_{k}}{{Precision}_{k-}{Recall}_{k}}\) The confusion matrix for the algorithm from the analysis shows exceptional proficiency in classifying the five cases of STDs. The diagonal dominance shows that the model has a strong classification capability with minimal misclassifications which demonstrates one of the best classification abilities to date through their effectiveness and accuracy. The dataset was analyzed using a logistic regression algorithm which illustrated the logistic regression algorithm when applied in the classification process. In evaluating the effectiveness of the Logistic Regression model on the dataset, the results were as follows: a correctness of 0.9621, a exactness of 0.9621, a recall of 1.0, and an AUC score of 0.5. Despite high accuracy, precision, and recall, the AUC score of 0.5 indicates that the algorithm is not suitable for this dataset and fails to effectively distinguish between the different cases of STD. The confusion matrix in Fig. 8 above reveals that the algorithm grouped all instances as not having STDs, failing to identify any cases of STDs. While the ROC result is illustrated in Fig. 9 . After evaluating all the models, the Gaussian Naïve Bayes algorithm emerged as the best performer for classifying instances of STDs. From the analysis, it seems that it achieved the highest AUC score of 0.7874 and displayed the most favourable ROC curve, with the greatest distance from the random classifier line. The Decision Tree classifier followed, with the second highest AUC score of 0.5843 for certain STD cases. Despite having high accuracy (0.9621), both the Support Vector Classifier and the Logistic Regression Classifier underperformed in classifying STDs due to their minimal AUC score of 0.5. CONCLUSION In the southwestern region of Nigeria, the spread of sexually transmitted diseases (STDs) among young individuals is continually increasing. This rise can largely be attributed to various factors, with lifestyle choices being the most significant. To address this issue, several classification algorithms were developed to predict whether an adolescent in southwestern Nigeria is likely to contract an STD based on their lifestyle choices. In this investigation, machine learning algorithms such as Support Vector Machine, Decision Tree, Gaussian Naïve Bayes and Logistic Regression were used. The models were evaluated using metrics such as precision, accuracy, confusion matrix, recall, AUC scores, and ROC curves to identify the most suitable algorithm for predicting STD occurrences. Hence, from the analysis results from the models, the following results were achieved are support vector classifier got an accuracy of 0.9621 and an AUC score of 0.5, the Decision Tree classifier recorded an exactness of 0.9393, correctness of 0.9685, a recall of 0.9685, and an AUC score of 0.5843, Gaussian Naïve Bayes classifier demonstrated a precision of 0.5909, a precision of 1.0, a recall of 0.5748, and the highest AUC score of 0.7874. while logistic regression had an exactness of 0.9621, an accuracy of 0.9621, and an AUC score of 0.5. Also, models such as feedforward multi-layer neural artificial networks, Naïve Bayes, Radial Basis Function Kernel (RBF), Polynomial Basis Kernel, SVM using, Extreme Gradient Boosting, Random Forest and Gradient Boosting Machine were considered in the analysis. Finally, based on the performance metrics, especially the AUC score, the Gaussian Naïve Bayes algorithm emerged as the best fit for the task having a high AUC score of 0.7874 thereby indicating the best ability to accurately predict the likelihood of STDs among young individuals in the region. Declarations Ethical Approval and consent to participate: Ethical approval of the study was obtained from the ethical committee of the Osun State Ministry of Health with the ethical approval number of OSHREC/PR5/569T/223 and participants’ assents and consent were obtained as applicable after the aims of the study was explained to them in clear and plain language before participating in the study. The study was performed in compliance with the Declaration of Helsinki of 1991 and its subsequent amendments on research involving human subjects following key principles of ethical research on recruitment, obtaining informed consent, minimal risk to participants, privacy and confidentiality of data, and transparency we also ensure that all participants were treated with respect and their health as well as their rights were protected. Availability of supporting data : Not applicable Funding: the author declared that they received no funding or gratification in the course of carrying out this research Author Contribution Conceptualization, ONE, OB, GO, COO and AAA; methodology, ONE, AOA, AAS and AAA; software, ONE, AAA, COO and AAS; validation, ONE, OB, GO, COO and AAA; formal analysis, ONE, GO, OB and COO; investigation, ONE, NOO, AOA and AAA; resources, ONE, COO, OB, GO and NOO; data curation, ONE, OB, GO, AAS and AAA; writing—original draft preparation, ONE, OB, COO, GO, AAA, AOA, AAS and NOO; writing—review and editing, ONE, OB, COO, AAA, GO, AOA, AAS and NOO; visualization, ONE, OB and GO; supervision, ONE and OB; project administration, ONE and OB; funding acquisition, ONE and COO. References A. Azizi; K. Rios-Soto; A. Mubayi, J.M. Hyman (2017) A risk-based model for predicting the impact of using condoms on the spread of sexually transmitted infections, The National Center for Biotechnology Information, KeAi Infectious Disease Modelling, 2(1)100–112. Centers for Disease Control and Prevention, STDs in Adolescents and Young Adult (2022): Sexually Transmitted Diseases Surveillance. Available online: https://www.cdc.gov/std/stats18/adolescents.htm (accessed on 18 February 2022) J. Daniels (2022) Why we should drive it home for the Nigerian youth, Vanguard Report, January 16, 2022. National bureau of statistics (2012). National Baseline Youth Survey. National Bureau of statistic, Abuja . N. K. R. Sri Preethaa, A. Muthuramalingam, Y. Natarajan, G. Wadhwa, A. A. Y. Ali (2023) A Comprehensive Review on Machine Learning Techniques for Forecasting Wind Flow Pattern, Sustainability 15(17)12914. O. N. Emuoyibofarhe, S. Adebayo, A. Ibitoye, M. O. Ayomide, T. Aderibigbe (2019) Predictive System for Heart Disease Using a Machine Learning Trained Model, International Journal of Computer (IJC) 34(1)140-152. A. Ogunleye, O. Adeoye, T. Adewale (2023) Development of a Machine Learning Model for the Prediction of Sexually Transmitted Diseases among the Youths in Southwestern Nigeria, International Journal of Infectious Diseases 78 (I)112-125. O.N. Emuoyibofarhe, R. F. Famutimi, D.O. Olanloye, S. Adebayo, A.A. Abdulraheem (2021) Development of Workforce Diseases Detection System Using Machine Learning Models. Journal of Applied Sciences, Information and Computing © School of Mathematics and Computing, Kampala International University,1(2)24–32. O. Adeboye, K.A. Bashiru, H. Afolabi, T. Ojurongbe(2023) Diagnosing Sexually Transmitted Disease from Some Symptoms Using Machine Learning Models, Journal of Statistical Modelling and Analytics, 5(1)65-80. X. Xu, C.K. Fairley, E.P.F Chow. et al. (2022) Using machine learning approaches to predict timely clinic attendance and the uptake of HIV/STI testing post clinic reminder messages. Sci Rep (12) 875. V.U. Nwadike, O. Olusanya, G.C. Anaedobe, I. Kalu, K. C. Ojide (2015)Patterns of sexually transmitted infections in patients presenting in special treatment clinic in Ibadan south western Nigeria, Pan African Medical Journal, (Published in partnership with the African Field Epidemiology Network (AFENET), 21(222) 1937- 8688. C. Nzoputam, V.Y. Adam, O. Nzoputam(2022) Knowledge, Prevalence and Factors Associated with Sexually Transmitted Diseases among Female Students of a Federal University in Southern Nigeria. Venereology, 1, 81–97. J. S. Kiran, J. Avanija, A. R. Reddy, G. N. R. Devi, N. S. Charan, F. Tabeen(2023) Early Prediction of Sepsis Utilizing Machine Learning Models in Evolution in Computational Intelligence (2023), Proceedings of the 11th International Conference on Frontiers of Intelligent Computing: Theory and Applications (FICTA 2023),319–327. H. R. Elder, S. Gruber, S. J. Willis, N. Cocoros , M. Callahan , E. W. Flagg , M. Klompas, K. K. Hsu (2021) Can Machine Learning Help Identify Patients at Risk for Recurrent Sexually Transmitted Infections? Sex Transm Dis PMID: 32810028 PMCID: PMC10949112,48(1):56-62. A. K. Kassaw, T. M. Yilma, Y. Sebastian, A. Y. Birhanu, M. S. Melaku, S. S. Jemal (2023) Spatial distribution and machine learning prediction of sexually transmitted infections and associated factors among sexually active men and women in Ethiopia, evidence from EDHS 2016, BMC Infectious Diseases, 23(49)1895. P.M. Latt, N.N. Soe, X. Xu, R. Rahman, E.P.F. Chow, J.J. Ong, C. Fairley, L. Zhang (2023) Assessing disparity in the distribution of HIV and sexually transmitted infections in Australia: a retrospective cross-sectional study using Gini coefficients. BMJ Public Health, 12, e000012. R.K. Barman, A.K. Chakrabarti, S. Dutta (2023) Prediction of Phage Virion Proteins Using Machine Learning Methods. Molecules, 28(5), 1–12. F. Nourimand, A. Keramat, M. Sayahi, L. Bozorgian, Z. Hashempour (2022)A systematic review on Machine Learning for Sexually Transmitted Infections (STIs), Diagnosis and Risk Prediction, Indian J Sex Transm Dis AIDS, 43(2)117–127. World Health Organization (2020) Sexually transmitted Infections. stat https://www.who.int/news-room/fact-sheets/detail/sexually-transmitted-infections-(stis). WHO. Sexually transmitted infections (STIs) 2019. https://www.health.ny.gov/diseases/communicable/std/. WHO. Sexually transmitted infections Europe: WHO; 2021. https://www.euro.who.int/en/health-topics/communicable-diseases/sexually-transmitted-infections/sexually-transmitted-infections. Adeyinka Oluwabusayo Abiodun, Kingsley Eghonghon Ukhurebor, Femi Alamu, Ibitoye Ayodeji, Adetoye Adeyemo, Ozichi N. Emuoyibofarhe, Lucky Evbuomwan, Oseremen Ebhote, Williams Omokhudu Odiwo, Grace Egenti, Adedoyin Abiodun Talabi (2024) Factors associated with poor perceptions of the COVID-19 pandemic in Africa. Journal of Infrastructure, Policy and Development. 8(8): 4770. https://doi.org/10.24294/jipd.v8i8.4770 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted 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-5404906","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":378430594,"identity":"b1d2d435-d13f-49da-ace6-02c0bf648c01","order_by":0,"name":"Ozichi N. Emuoyibofarhe","email":"data:image/png;base64,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","orcid":"","institution":"Bowen University","correspondingAuthor":true,"prefix":"","firstName":"Ozichi","middleName":"N.","lastName":"Emuoyibofarhe","suffix":""},{"id":378430595,"identity":"fdfd8f8c-6853-4007-a893-66d53d46c488","order_by":1,"name":"Olubayode. Bamidele","email":"","orcid":"","institution":"Bowen University","correspondingAuthor":false,"prefix":"","firstName":"Olubayode.","middleName":"","lastName":"Bamidele","suffix":""},{"id":378430596,"identity":"a0e7ad34-7ce7-402f-ad61-c52989ed357c","order_by":2,"name":"Christian O. Osueke","email":"","orcid":"","institution":"Bowen University","correspondingAuthor":false,"prefix":"","firstName":"Christian","middleName":"O.","lastName":"Osueke","suffix":""},{"id":378430597,"identity":"85a9e2a6-b425-4090-972d-0b27e5e93f60","order_by":3,"name":"Adetoye A. Adeyemo","email":"","orcid":"","institution":"Bowen University","correspondingAuthor":false,"prefix":"","firstName":"Adetoye","middleName":"A.","lastName":"Adeyemo","suffix":""},{"id":378430598,"identity":"5b6d41f8-ae17-4bb4-869d-a514d3fa9669","order_by":4,"name":"Gideon Ojo","email":"","orcid":"","institution":"Bowen University","correspondingAuthor":false,"prefix":"","firstName":"Gideon","middleName":"","lastName":"Ojo","suffix":""},{"id":378430599,"identity":"88c6ba5a-c192-4652-8dcd-f6a9f10dded3","order_by":5,"name":"Abiodun O. Adeyinka","email":"","orcid":"","institution":"National Open University of Nigeria","correspondingAuthor":false,"prefix":"","firstName":"Abiodun","middleName":"O.","lastName":"Adeyinka","suffix":""},{"id":378430600,"identity":"2594eb4e-f760-4c6f-9a60-d190cbdea80b","order_by":6,"name":"Abraham A. Sunday","email":"","orcid":"","institution":"Bowen University","correspondingAuthor":false,"prefix":"","firstName":"Abraham","middleName":"A.","lastName":"Sunday","suffix":""},{"id":378430601,"identity":"cf20a8b0-797d-4585-84b8-5b4b062b25f9","order_by":7,"name":"Ngozi O. Osueke","email":"","orcid":"","institution":"Bowen University","correspondingAuthor":false,"prefix":"","firstName":"Ngozi","middleName":"O.","lastName":"Osueke","suffix":""}],"badges":[],"createdAt":"2024-11-06 18:23:26","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5404906/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5404906/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":71735797,"identity":"7767412a-fea9-4a20-be4d-edbe2e082baa","added_by":"auto","created_at":"2024-12-18 07:30:44","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":285969,"visible":true,"origin":"","legend":"\u003cp\u003eUn-cleaned Dataset Sample\u003c/p\u003e","description":"","filename":"image1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-5404906/v1/dd4546fdea710da80c32c654.jpeg"},{"id":71734472,"identity":"007d24c3-d545-4b87-90c9-a46f9c97dab2","added_by":"auto","created_at":"2024-12-18 07:22:44","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":163091,"visible":true,"origin":"","legend":"\u003cp\u003eLabel encoding process\u003c/p\u003e","description":"","filename":"image2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-5404906/v1/1cd4055b428e436d17dae94d.jpeg"},{"id":71734468,"identity":"6faa5a51-9e0e-4478-bb8a-eea150a860be","added_by":"auto","created_at":"2024-12-18 07:22:44","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":29643,"visible":true,"origin":"","legend":"\u003cp\u003eConfusion Matrix of the SVM Classification Algorithm\u003c/p\u003e","description":"","filename":"image3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-5404906/v1/716ba53eac8a9d5c81b2a5ff.jpeg"},{"id":71735795,"identity":"409133fe-3d4e-4f43-b977-46ef025b8cd4","added_by":"auto","created_at":"2024-12-18 07:30:44","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":44923,"visible":true,"origin":"","legend":"\u003cp\u003eConfusion matrix of the Decision Tree Algorithm\u003c/p\u003e","description":"","filename":"image4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-5404906/v1/72fb9a80e4c542edc757a592.jpeg"},{"id":71735798,"identity":"904e7b36-b2c2-4aff-91dc-3cc2f1f19d07","added_by":"auto","created_at":"2024-12-18 07:30:44","extension":"jpeg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":44581,"visible":true,"origin":"","legend":"\u003cp\u003eROC curve of the decision tree classifier\u003c/p\u003e","description":"","filename":"image5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-5404906/v1/c6c66552f750dafa2f9cd138.jpeg"},{"id":71734474,"identity":"9b9ad282-d2bf-4aef-bb5d-f69abdb0aa3f","added_by":"auto","created_at":"2024-12-18 07:22:44","extension":"jpeg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":46150,"visible":true,"origin":"","legend":"\u003cp\u003eConfusion Matrix of the GNB classifier\u003c/p\u003e","description":"","filename":"image6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-5404906/v1/63656ccfaa521cd0fb7f3038.jpeg"},{"id":71734470,"identity":"aece8900-b406-44a5-86c9-3f8990853e11","added_by":"auto","created_at":"2024-12-18 07:22:44","extension":"jpeg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":42626,"visible":true,"origin":"","legend":"\u003cp\u003eThis illustrates the ROC curve of the algorithm.\u003c/p\u003e","description":"","filename":"image7.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-5404906/v1/cd47c9346166f7f5651de78b.jpeg"},{"id":71735796,"identity":"2d2e864e-69f5-4efc-b8ed-02d95d7ac616","added_by":"auto","created_at":"2024-12-18 07:30:44","extension":"jpeg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":51547,"visible":true,"origin":"","legend":"\u003cp\u003eConfusion Matrix of the Algorithm\u003c/p\u003e","description":"","filename":"image8.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-5404906/v1/16684ece37058d450fdd427b.jpeg"},{"id":71737107,"identity":"458cb0d7-2a2b-4f0a-bd5a-7be0bb69a98b","added_by":"auto","created_at":"2024-12-18 07:38:45","extension":"jpeg","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":42734,"visible":true,"origin":"","legend":"\u003cp\u003eROC curve of the Logistic Regression Model\u003c/p\u003e","description":"","filename":"image9.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-5404906/v1/19ddbe4ba7b7cdc6481a9d85.jpeg"},{"id":74336188,"identity":"1963e11a-6481-4f14-8876-8f5359eaa6bc","added_by":"auto","created_at":"2025-01-21 07:40:02","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1261684,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5404906/v1/5b7e9040-4db9-4362-aa36-860765ce995a.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"A Machine Learning Model for the Prediction of Sexually Transmitted Diseases among the Youths in Southwestern Nigeria","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eSexually transmitted diseases (STDs) are infections spread from infected individuals to other through unprotected sex. It is either a curable or incurable disease or infection that is primarily transmitted through unprotected sexual contact. Syphilis, gonorrhea, and Chlamydia Trachomatis are curable sexually transmitted infections (STIs), while herpes simplex, hepatitis B, human papillomavirus (HPV), and human immune virus (HIV) are incurable yet modifiable STIs, Azizi et al. [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Sexually transmitted diseases have been a persistent issue in Nigeria for as long as anybody can remember, and they do not appear to be going away anytime soon. In recent times, several sexually transmitted diseases (STDs) have been globally spread, which are majorly asymptomatic, Centers for Disease and Prevention [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Although there are no official statistics available, previous research has indicated that the prevalence of treatable STIs in Nigeria ranges from 0\u0026ndash;18%, Daniel [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Though Western education is widely available in Nigeria's southwest states, the youths are either ignorant of the risks associated with unprotected sexual activity or prefer to ignore them. Additionally, research has indicated that youth often have a stronger propensity to engage in risky sexual behaviour with several partners due to their desire to explore their sexualities. As such, they are likely to have infectious diseases due to their careless attitude. STDs, the bulk of which are asymptomatic, are contracted daily around the world, ] National bureau of statistics [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Nigeria, being a highly populated black race, has about 200\u0026nbsp;million individuals with about 2.61% yearly birth rate increment, of which 16.8% are aged 15 to 35 grouped as youths, Centers for Disease and Prevention [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. According to the Federal Ministry of Youth Development and the National Bureau of Statistics, the age group 15 to 35 years was appraised to be over 64\u0026nbsp;million individuals in 2012 (the most recent census).\u003c/p\u003e \u003cp\u003eIn the same census, it was estimated that the number of youths in the southwestern states of Nigeria was 13\u0026nbsp;million individuals, accounting for approximately 20% of the total population of youths in Nigeria, Daniel [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Therefore, the southwest states in Nigeria are Osun, Oyo, Ekiti, Ondo, and Ogun, which are the focus of this study. Machine learning approaches are widely used in the analysis and prediction of many different facets of human behaviour, Sri Preethaa et al [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. The scientific investigation of mathematical theories and techniques applied to computing devices to carry out specified tasks without utilizing explicit instructions while concentrating on patterns and inferences is known as machine learning (ML), Emuoyibofarhe et al. [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Machine learning algorithms are employed to forecast the risk of sexual transmission among youths; and carry out certain task, which focuses on patterns and inferences rather than using explicit instructions, Ogunleye et al. [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. A machine learning technique for the analysis of data computerizes the formulation of models. Artificial intelligence is created to make machines learn from data, spot patterns, and conclude with little human effort, Emuoyibofarhe et al. [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. To create a machine learning algorithm, output predicted functions must be identified. As a result, this research aims to predict the sexual infectious disease rate among youths in southwestern Nigeria using a machine learning model, with the scope of the study limited to the prediction of STDs among adolescents. A machine learning tool-based model is developed to predict the sexual infections and HIV risk for one year. Utilizing models such as random forests, regression methods, support vector machines, and bagging ensembles, the tool analyzed data collected from March 2, 2015, to December 31, 2019. The final analysis reported the following incidence rates per 100 person-years: 16.95 [95% CI: 16.82\u0026ndash;17.67] for gonorrhea, 17.96 [95% CI: 16.79\u0026ndash;18.13] for chlamydia, and 0.21 [95% CI: 0.17\u0026ndash;0.27] for syphilis and HIV, Adeboye et al. [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. In a separate study, referenced in Xu et al. [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e], machine learning techniques were employed to forecast testing HIV/STI and clinic attendance in response to clinical reminder messages. The study utilized data from 3,044 consultations to develop and refine the predictive models. A range of algorithms was implemented, including feedforward multi-layer neural artificial networks, generalized Bayesian linear models K-nearest neighbours, linear vector support machines, Naive Bayes, polynomial basis kernel, extreme gradient boosting, random forest, gradient boosting machines, and elastic net, ridge The\u003c/p\u003e \u003cp\u003emedian age of participants was 31.0 years, with reminder messages distributed via email (49.6%), SMS (30.6%), and other methods (19.8%). The study found that 15.5% of participants lived with HIV, and 19.1% had the symptoms of STI at the time of their clinic visit. Notably, 29.5% of patients attended the clinic within 30 days of the reminder.\u003c/p\u003e \u003cp\u003eIn Azizi et al. [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e], the researchers developed a predictive risk-based model to analyze the impact of condoms on the sexual disease transmission rate. The results reveal that for a random mixing, the rising use of condoms will effectively decline the disease spread pattern in the population. Hence the model computed the risk of infection increase for individuals with many sexual partners, and the need for people with many sexual partners to use condoms for infection risk reduction. Nwadike et al. [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e] undertook a study where records of patients diagnosed with sexually\u003c/p\u003e \u003cp\u003etransmitted diseases were analyzed to recognize patterns of contraction of the infections within the patients. The records of 506 patients were used where females are 56.3% (285) and males are 43.7% (221) and their age ranges from one to eighty. In the analysis, the 1\u0026ndash;10 age group and 71\u0026ndash;80 age group seemed to be the lowest age group representation when the various features such as education level, sex, age, yeast cells presence, and disease venereal laboratory research test were used during the test.\u003c/p\u003e \u003cp\u003eFurthermore, in Nzoputam et al. [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e], the researchers stipulated the feature model essential chart that shows vagina itching and vagina discharge as the highest impact level on the patient diagnosed possibility with STDs, where 100% logistic regression shows a properly predicted model all the 101 true positives and 309 true negatives to have been wrongly diagnosed to be zero. According to Kiran et al. [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e], Sepsis infection as the name implies is a risky infection initiated by the individual\u0026rsquo;s rejoinder to a disease, which normally brings about organ failure, damage of tissue, or death due to inflammation spreading all over the body. The results obtained by the researcher show that the machine learning algorithms give accurate prediction of sepsis with high specificity and sensitivity thereby giving an auspicious solution for primary clinical detection of sepsis. This study proposed a possible machine learning model for enhancing sepsis control and detection providing a basis for future study.\u003c/p\u003e \u003cp\u003eAlso, a study in Elder et al. [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e] was carried out using a machine learning model to predict the risk of sexual infection among the youths in Southwestern Nigeria; in the study, data from a cross-sectional inspection of 2,000 people with age ranges from 15 to 24years were utilized to explore STD risk factors. Various machine learning algorithms, including vector support machines, decision trees, and logistic regression were employed for data analysis and to develop an analytical model. The resulting model, which incorporated features such as age, gender, sexual behaviour, and socioeconomic status, demonstrated high accuracy in predicting STD risk factors. The findings indicate that machine learning techniques hold significant potential for early detection and prevention of STDs among youth particularly in the resource-limited settings.\u003c/p\u003e \u003cp\u003eA study conducted in Ethiopia applied machine learning techniques to identify predictors of sexually transmitted infections (STIs) and analyzed their geographic distribution across various regions. By understanding these predictors and their spatial patterns, policymakers can gain deeper insights into STI issues and tailor interventions more effectively. This research highlights the potential of machine learning in forecasting STIs within Ethiopia, Kassaw et al. [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Also, the study in Latt et al. [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e] examined the potential of electronic health records for data routine to predict patients with a newly diagnosed sexually transmitted infection (STI) are most likely to contract another STI within the following 1 to 2 years. Furthermore, the study in Barman et al. [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e] analyzed the disparity in the HIV distribution and sexually spread disease among various groups of populaces in Australia. Using Gini coefficients, the research found significant variations in HIV/STI risk, with higher Gini scores approaching one indicating greater inequality. Advancements in technology and the widespread use of the Internet, particularly among young people, have created valuable opportunities for leveraging eHealth in disease prevention. Nourimand et al. [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. eHealth approaches can play a significant role in preventing sexually transmitted diseases (STDs), especially among the young demographic who frequently engage with new technologies and are at increased risk for STDs. Our systematic review specifically explored the preventative potential of eHealth for sexually transmitted infections (STIs), highlighting its various benefits and applications, WHO [\u003cspan additionalcitationids=\"CR20\" citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. Also, a study stipulated that if the poor attitude or perception towards complying with the preventive measures continues, COVID-19 cases in Africa can result in an increase beyond the current spread. Adeyinka et al. [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]\u003c/p\u003e"},{"header":"RESEARCH METHODOLOGY","content":"\u003cp\u003eThe research methodology adopted for the machine learning model for the prediction of sexually transmitted diseases among the youths in Southwestern Nigeria are respectively given.\u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Model Development\u003c/h2\u003e \u003cp\u003eA total of six machine learning models were executed on the dataset and the most appropriate model is chosen based on the assessment process. The details of how the entire process was done is detailed below:\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003e2.1.1 Data Collection\u003c/h3\u003e\n\u003cp\u003ePrimary data was the main and only source of data for the machine learning process. The data collection was done using the questionnaire data collection process via Google forms, which is a useful platform adopted to create and share digital survey for the research work. The platform had a function to convert the responses for easy analysis of the CSV format. A total of 529 responses were used to create the dataset for this study. A preview of the un-cleaned sample dataset from the google form is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e2.1.2 Data Loading and Analysis\u003c/p\u003e \u003cp\u003eFeature Extraction\u003c/p\u003e \u003cp\u003eThe column containing the type of STD contracted, if the respondent has ever had an STD, is excluded because the dataset's output label is the STD column, and the specific type of STD is not considered a relevant feature for predicting the output label. Hence, the process of extracting this feature from the dataset is as shown below.\u003c/p\u003e \u003cp\u003eData Preprocessing\u003c/p\u003e \u003cp\u003eDuring this stage, data cleaning was conducted, and string attributes were encoded to make them compatible with the algorithm. Microsoft excel facilitated the identification and replacement of these attributes. Subsequently, all string entities underwent label encoding, converting them into categorical numbers suitable for algorithmic processing. The preprocessing process is shown in the figure below:\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAbove in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, all features were encoded. The outcome of this process reveals the newly cleaned dataset presented below.\u003c/p\u003e \u003cp\u003eImplementation and Evaluation of Classification Algorithms\u003c/p\u003e \u003cp\u003eThe sanitized dataset is divided into two training dataset sections, 75% of the data serves as the training set, while the remaining 25% serves as the testing set. Columns one through seventeenth are designated as input labels, whereas the eighteenth column, represents the STD status which is designated as the output label. The training and testing data were utilized across all learning algorithms in subsequent processes using the Support Vector Classification (SVC) method. The implementation of the SVC algorithm yielded a model with an accuracy of 0.96212 and an AUC score of 0.5. The corresponding confusion matrix is also carried out.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe confusion matrix in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e indicates that the algorithm successfully classified all instances where individuals did not have the disease but failed to correctly identify any cases where individuals did have the disease. The AUC score of 0.5 further demonstrates the algorithm's poor performance, making it unsuitable for this classification problem.\u003c/p\u003e \u003cp\u003e3.0. Decision Tree Classification\u003c/p\u003e \u003cp\u003eThis is a method that is used to diagnose Sexually Transmitted Diseases (STDs) by analyzing the patient data such as the symptoms and test results. This is done by actually splitting the data at each node which is based on the availability of the specific features thereby leading to a final classification at the leaf nodes which in turn helps to identify the particular STD or stating that there is no infection. The method is straightforward, easy to interpret and highlights and also provides a quick and effective way to assist healthcare providers in the diagnostics of STDs.\u003c/p\u003e \u003cp\u003eModel Evaluation\u003c/p\u003e \u003cp\u003eThe implemented algorithm achieved a correctness of 0.9393, an AUC score of 0.5843, an exactness of 0.9685, and a recall of 0.9685. The confusion matrix for this model is examined as presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe confusion matrix revealed that the algorithm successfully classified 123 negative STD cases but correctly identified only one out of five positive STD cases. With an AUC of 0.5843, the model demonstrates limited performance. The ROC curve for this algorithm is illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e below.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e4.0 Gaussian Na\u0026iuml;ve Bayers Algorithm\u003c/p\u003e \u003cp\u003eThis method is particularly useful as it can incorporate local data on STD prevalence and common symptoms and is used to diagnose Sexually Transmitted Diseases (STDs) in this work by leveraging patient data with the underlined patient features based on the symptoms. Hence, the model attained an accuracy of 0.5909, a correctness of 1.0, a recall of 0.5748, and an AUC score of 0.7874. The AUC score of 0.7874 indicates that the algorithm is a stronger classifier than other algorithms when compared. The misperception matrix of the algorithm is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e below and the ROC curve for the algorithm is illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe confusion matrix for the algorithm classifying the five types of STDs can be represented using the analysis below:\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"No\" id=\"Taba\" border=\"1\"\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\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eActual/Predicted\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSTD\u0026nbsp;1\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSTD\u0026nbsp;2\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSTD\u0026nbsp;3\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSTD\u0026nbsp;4\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSTD\u0026nbsp;1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eC12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eC13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eC14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eC15\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSTD\u0026nbsp;2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eC22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eC23\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eC24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eC25\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSTD\u0026nbsp;3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC31\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eC32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eC33\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eC34\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eC35\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSTD\u0026nbsp;4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC41\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eC42\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eC43\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eC44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eC45\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSTD\u0026nbsp;5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC51\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eC52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eC53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eC54\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eC55\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e\n\u003ch3\u003eInterpretation\u003c/h3\u003e\n\u003cp\u003eDiagonal Elements (Cii​): These represent the correctly classified instances for each class. For example, C11​ is the number of correctly predicted cases of STD 1. Off-Diagonal Elements (Cij where I\u0026thinsp;\u0026ne;\u0026thinsp;j): These represent the misclassified instances. For instance, C12 is the number of STD 1 cases that were incorrectly predicted as STD 2.\u003c/p\u003e \u003cp\u003ePerformance Analysis\u003c/p\u003e \u003cp\u003eFrom the confusion matrix, we can derive various performance metrics to assess the classification ability of the algorithm. Common metrics include:\u003c/p\u003e \u003cp\u003e \u003cb\u003eAccuracy\u003c/b\u003e: The overall proportion of correctly classified instances.\u003c/p\u003e \u003cp\u003eAccuracy = \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\frac{\\sum\\:_{i=1}^{N}C11}{\\sum\\:_{i=1\\:}^{N}\\sum\\:_{j=1}^{N}Cij}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e \u003cb\u003ePrecision for Class\u003c/b\u003e k: The correct proportion of the predicted occurrences of class k out of all occurrences foreseen as class k.\u003c/p\u003e \u003cp\u003ePrecision = ​​\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\frac{{C}_{KK}}{\\sum\\:_{j=1}^{N}{C}_{ij}}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e \u003cb\u003eRecall for Class\u003c/b\u003e k: The correct quantity of the forecast instances of class k out of all definite instances of class k.\u003c/p\u003e \u003cp\u003eRecall = \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\frac{{C}_{kk}}{\\sum\\:_{J=1}^{N}{C}_{Kj}}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003e \u003cb\u003eF1-Score for class\u003c/b\u003e k: The harmonic means of exactness and recall for class k.\u003c/p\u003e \u003cp\u003eF\u003csub\u003e1 \u0026ndash;\u003c/sub\u003e Score\u003csub\u003ek\u003c/sub\u003e = \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\frac{{Precision}_{k}-\\:{Recall}_{k}}{{Precision}_{k-}{Recall}_{k}}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003eThe confusion matrix for the algorithm from the analysis shows exceptional proficiency in classifying the five cases of STDs. The diagonal dominance shows that the model has a strong classification capability with minimal misclassifications which demonstrates one of the best classification abilities to date through their effectiveness and accuracy.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe dataset was analyzed using a logistic regression algorithm which illustrated the logistic regression algorithm when applied in the classification process. In evaluating the effectiveness of the Logistic Regression model on the dataset, the results were as follows: a correctness of 0.9621, a exactness of 0.9621, a recall of 1.0, and an AUC score of 0.5. Despite high accuracy, precision, and recall, the AUC score of 0.5 indicates that the algorithm is not suitable for this dataset and fails to effectively distinguish between the different cases of STD.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe confusion matrix in Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e above reveals that the algorithm grouped all instances as not having STDs, failing to identify any cases of STDs. While the ROC result is illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAfter evaluating all the models, the Gaussian Na\u0026iuml;ve Bayes algorithm emerged as the best performer for classifying instances of STDs. From the analysis, it seems that it achieved the highest AUC score of 0.7874 and displayed the most favourable ROC curve, with the greatest distance from the random classifier line. The Decision Tree classifier followed, with the second highest AUC score of 0.5843 for certain STD cases. Despite having high accuracy (0.9621), both the Support Vector Classifier and the Logistic Regression Classifier underperformed in classifying STDs due to their minimal AUC score of 0.5.\u003c/p\u003e"},{"header":"CONCLUSION","content":"\u003cp\u003eIn the southwestern region of Nigeria, the spread of sexually transmitted diseases (STDs) among young individuals is continually increasing. This rise can largely be attributed to various factors, with lifestyle choices being the most significant. To address this issue, several classification algorithms were developed to predict whether an adolescent in southwestern Nigeria is likely to contract an STD based on their lifestyle choices. In this investigation, machine learning algorithms such as Support Vector Machine, Decision Tree, Gaussian Na\u0026iuml;ve Bayes and Logistic Regression were used. The models were evaluated using metrics such as precision, accuracy, confusion matrix, recall, AUC scores, and ROC curves to identify the most suitable algorithm for predicting STD occurrences. Hence, from the analysis results from the models, the following results were achieved are support vector classifier got an accuracy of 0.9621 and an AUC score of 0.5, the Decision Tree classifier recorded an exactness of 0.9393, correctness of 0.9685, a recall of 0.9685, and an AUC score of 0.5843, Gaussian Na\u0026iuml;ve Bayes classifier demonstrated a precision of 0.5909, a precision of 1.0, a recall of 0.5748, and the highest AUC score of 0.7874. while logistic regression had an exactness of 0.9621, an accuracy of 0.9621, and an AUC score of 0.5. Also, models such as feedforward multi-layer neural artificial networks, Na\u0026iuml;ve Bayes, Radial Basis Function Kernel (RBF), Polynomial Basis Kernel, SVM using, Extreme Gradient Boosting, Random Forest and Gradient Boosting Machine were considered in the analysis. Finally, based on the performance metrics, especially the AUC score, the Gaussian Na\u0026iuml;ve Bayes algorithm emerged as the best fit for the task having a high AUC score of 0.7874 thereby indicating the best ability to accurately predict the likelihood of STDs among young individuals in the region.\u003c/p\u003e "},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthical Approval\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eand consent to participate:\u003c/strong\u003e Ethical approval of the study was obtained from the ethical committee of the Osun State Ministry of Health with the ethical approval number of OSHREC/PR5/569T/223 and participants\u0026rsquo; assents and consent were obtained as applicable after the aims of the study was explained to them in clear and plain language before participating in the study. The study was performed in compliance with the Declaration of Helsinki of 1991 and its subsequent amendments on research involving human subjects following key principles of ethical research on recruitment, obtaining informed consent, minimal risk to participants, privacy and confidentiality of data, and transparency we also ensure that all participants were treated with respect and their health as well as their rights were protected.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of supporting data\u003c/strong\u003e: Not applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding:\u003c/strong\u003e the author declared that they received no funding or gratification in the course of carrying out this research\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contribution\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConceptualization, ONE, OB, GO, COO and AAA; methodology, ONE, AOA, AAS and AAA; software, ONE, AAA, COO and AAS; validation, ONE, OB, GO, COO and AAA; formal analysis, ONE, GO, OB and COO; investigation, ONE, NOO, AOA and AAA; resources, ONE, COO, OB, GO and NOO; data curation, ONE, OB, GO, AAS and AAA; writing\u0026mdash;original draft preparation, ONE, OB, COO, GO, AAA, AOA, AAS and NOO; writing\u0026mdash;review and editing, ONE, OB, COO, AAA, GO, AOA, AAS and NOO; visualization, ONE, OB and GO; supervision, ONE and OB; project administration, ONE and OB; funding acquisition, ONE and COO.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eA. Azizi; K. Rios-Soto; A. Mubayi, J.M. Hyman (2017) A risk-based model for predicting the impact of using condoms on the spread of sexually transmitted infections, The National Center for Biotechnology Information, KeAi Infectious Disease Modelling, 2(1)100\u0026ndash;112.\u003c/li\u003e\n \u003cli\u003eCenters for Disease Control and Prevention, STDs in Adolescents and Young Adult (2022): Sexually Transmitted Diseases Surveillance. Available online: https://www.cdc.gov/std/stats18/adolescents.htm (accessed on 18 February 2022)\u003c/li\u003e\n \u003cli\u003eJ. Daniels (2022) Why we should drive it home for the Nigerian youth, Vanguard Report, January 16, 2022.\u003c/li\u003e\n \u003cli\u003eNational bureau of statistics (2012). National Baseline Youth Survey. National Bureau of statistic, Abuja\u003cem\u003e.\u003c/em\u003e\u003c/li\u003e\n \u003cli\u003eN. K. R. Sri Preethaa, A. Muthuramalingam, Y. Natarajan, G. Wadhwa, A. A. Y. Ali (2023) A Comprehensive Review on Machine Learning Techniques for Forecasting Wind Flow Pattern, Sustainability 15(17)12914.\u003c/li\u003e\n \u003cli\u003eO. N. Emuoyibofarhe, S. Adebayo, A. Ibitoye, M. O. Ayomide, T. Aderibigbe (2019) Predictive System for Heart Disease Using a Machine Learning Trained Model, International Journal of Computer (IJC) 34(1)140-152.\u003c/li\u003e\n \u003cli\u003eA. Ogunleye, O. Adeoye, T. Adewale (2023) Development of a Machine Learning Model for the Prediction of Sexually Transmitted Diseases among the Youths in Southwestern Nigeria, International Journal of Infectious Diseases 78 (I)112-125.\u003c/li\u003e\n \u003cli\u003eO.N. Emuoyibofarhe, R. F. Famutimi, D.O. Olanloye, S. Adebayo, A.A. Abdulraheem (2021) Development of Workforce Diseases Detection System Using Machine Learning Models. Journal of Applied Sciences, Information and Computing \u0026copy; School of Mathematics and Computing, Kampala International University,1(2)24\u0026ndash;32.\u003c/li\u003e\n \u003cli\u003eO. Adeboye, K.A. Bashiru, H. Afolabi, T. Ojurongbe(2023) Diagnosing Sexually Transmitted Disease from Some Symptoms Using Machine Learning Models, Journal of Statistical Modelling and Analytics, 5(1)65-80.\u003c/li\u003e\n \u003cli\u003eX. Xu, C.K. Fairley, E.P.F Chow. et al. (2022) Using machine learning approaches to predict timely clinic attendance and the uptake of HIV/STI testing post clinic reminder messages. Sci Rep (12) 875.\u003c/li\u003e\n \u003cli\u003eV.U. Nwadike, O. Olusanya,\u003csup\u003e\u0026nbsp;\u003c/sup\u003eG.C. Anaedobe, I. Kalu, K. C. Ojide (2015)Patterns of sexually transmitted infections in patients presenting in special treatment clinic in Ibadan south western Nigeria, Pan African Medical Journal, (Published in partnership with the African Field Epidemiology Network (AFENET), 21(222) 1937- 8688.\u003c/li\u003e\n \u003cli\u003eC. Nzoputam, V.Y. Adam, O. Nzoputam(2022) Knowledge, Prevalence and Factors Associated with Sexually Transmitted Diseases among Female Students of a Federal University in Southern Nigeria. Venereology, 1, 81\u0026ndash;97.\u003c/li\u003e\n \u003cli\u003eJ. S. Kiran, J. Avanija, A. R. Reddy, G. N. R. Devi, N. S. Charan, F. Tabeen(2023) Early Prediction of Sepsis Utilizing Machine Learning Models in Evolution in Computational Intelligence (2023), Proceedings of the 11th International Conference on Frontiers of Intelligent Computing: Theory and Applications (FICTA 2023),319\u0026ndash;327.\u003c/li\u003e\n \u003cli\u003eH. R. Elder, S. Gruber, S. J. Willis, N. Cocoros\u003csup\u003e\u0026nbsp;\u003c/sup\u003e, M. Callahan\u003csup\u003e\u0026nbsp;\u003c/sup\u003e, E. W. Flagg\u003csup\u003e\u0026nbsp;\u003c/sup\u003e, M. Klompas, K. K. Hsu (2021) Can Machine Learning Help Identify Patients at Risk for Recurrent Sexually Transmitted Infections? Sex Transm Dis PMID: 32810028 PMCID: PMC10949112,48(1):56-62.\u003c/li\u003e\n \u003cli\u003eA. K. Kassaw, T. M. Yilma, Y. Sebastian, A. Y. Birhanu, M. S. Melaku, S. S. Jemal (2023) Spatial distribution and machine learning prediction of sexually transmitted infections and associated factors among sexually active men and women in Ethiopia, evidence from EDHS 2016, BMC Infectious Diseases, 23(49)1895.\u003c/li\u003e\n \u003cli\u003eP.M. Latt, N.N. Soe, X. Xu, R. Rahman, E.P.F. Chow, J.J. Ong, C. Fairley, L. Zhang (2023) Assessing disparity in the distribution of HIV and sexually transmitted infections in Australia: a retrospective cross-sectional study using Gini coefficients. BMJ Public Health, 12, e000012.\u003c/li\u003e\n \u003cli\u003eR.K. Barman, A.K. Chakrabarti, S. Dutta (2023) Prediction of Phage Virion Proteins Using Machine Learning Methods. Molecules, 28(5), 1\u0026ndash;12.\u003c/li\u003e\n \u003cli\u003eF. Nourimand, A. Keramat, M. Sayahi, L. Bozorgian, Z. Hashempour (2022)A systematic review on Machine Learning for Sexually Transmitted Infections (STIs), Diagnosis and Risk Prediction, Indian J Sex Transm Dis AIDS, 43(2)117\u0026ndash;127.\u003c/li\u003e\n \u003cli\u003eWorld Health Organization (2020) Sexually transmitted Infections. stat https://www.who.int/news-room/fact-sheets/detail/sexually-transmitted-infections-(stis).\u003c/li\u003e\n \u003cli\u003eWHO. Sexually transmitted infections (STIs) 2019. https://www.health.ny.gov/diseases/communicable/std/.\u003c/li\u003e\n \u003cli\u003eWHO. Sexually transmitted infections Europe: WHO; 2021. https://www.euro.who.int/en/health-topics/communicable-diseases/sexually-transmitted-infections/sexually-transmitted-infections.\u003c/li\u003e\n \u003cli\u003eAdeyinka Oluwabusayo Abiodun, Kingsley Eghonghon Ukhurebor, Femi Alamu, Ibitoye Ayodeji, Adetoye Adeyemo, Ozichi N. Emuoyibofarhe, Lucky Evbuomwan, Oseremen Ebhote, Williams Omokhudu Odiwo, Grace Egenti, Adedoyin Abiodun Talabi (2024) Factors associated with poor perceptions of the COVID-19 pandemic in Africa. Journal of Infrastructure, Policy and Development. 8(8): 4770. https://doi.org/10.24294/jipd.v8i8.4770\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"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":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Machine learning model, Sexually transmitted diseases, Gaussian Naïve Bayers, Logistic regression","lastPublishedDoi":"10.21203/rs.3.rs-5404906/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5404906/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSexually transmitted diseases (STDs) are diseases which are spread between individuals through unprotected sexual contact. The spread has become rampant, especially among the youths nowadays who display promiscuous characteristics, which leads to a faster rate of the spread of the disease among the youths. Thus, this study aims to develop a machine learning model for an accurate analysis and prediction of the transmission rate of STDs among the youth within the southwestern region of Nigeria. For an approximate and optimize study, a questionnaire in Google form was administered to harvest opinions of youths within the stated demographic with respect to their health status, disease awareness, lifestyle choices and other characteristics. The collected primary dataset of 529 individual responses was used to build the machine learning model. The dataset was converted to comma-separated values (CSV) format to be trained and tested for a well-supervised machine learning model. Of the data collected, 75% served as training data and 25% served as testing data. Feature extraction, data visualization and data preprocessing were done to convert raw data into suitable machine learning. Taking from the results, a decision tree accuracy of 0.9393 with an area under the curve (AUC) score of 0.5843, logistic regression accuracy of 0.9621 with AUC score of 0.5, support vector machine accuracy of 0.96212 with AUC score of 0.5 and Gaussian Na\u0026iuml;ve Bayers machine learning algorithms accuracy score of 0.5909 with AUC score of 0.7874 were obtained. Hence, the Gaussian Na\u0026iuml;ve Bayers gave the best outcomes with an area under the curve (AUC) score of 0.79 and was able to correctly classify all 5 cases of STDs within the test set as compared to other algorithms.\u003c/p\u003e","manuscriptTitle":"A Machine Learning Model for the Prediction of Sexually Transmitted Diseases among the Youths in Southwestern Nigeria","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-12-18 07:22:39","doi":"10.21203/rs.3.rs-5404906/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"d8ef0080-5c84-4d9f-a8f4-89681464c6c1","owner":[],"postedDate":"December 18th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-01-21T07:39:21+00:00","versionOfRecord":[],"versionCreatedAt":"2024-12-18 07:22:39","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5404906","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5404906","identity":"rs-5404906","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}