Word Sense Disambiguation (WSD) in Indonesian Sentences Using Simplified Lesk Algorithm

Word Sense Disambiguation (WSD) in Indonesian Sentences Using Simplified Lesk Algorithm | 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 Word Sense Disambiguation (WSD) in Indonesian Sentences Using Simplified Lesk Algorithm Nurul Akhni, Abdiansah, Danny Matthew Saputra This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7472904/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 The Indonesian language contains several words with inherent ambiguity, meaning they possess more than one possible interpretation. Word Sense Disambiguation (WSD), a branch of Natural Language Processing (NLP), deals with the challenge of resolving this ambiguity and identifying the precise meaning of a word based on its context. Among the algorithms used for WSD, the Simplified Lesk algorithm stands out as particularly popular. To assess its effectiveness, tests were conducted using the Kamus Besar Bahasa Indonesia (KBBI) as a reference for word definitions, and a dataset of 300 Indonesian sentences containing ambiguous words and their respective meanings as determined by human perception. The research reveals that the configuration of the preprocessing phase plays a crucial role in accurately identifying the intended meaning. After evaluation, the overall accuracy achieved was 58% for the dataset, incorporating preprocessing techniques such as stemming and stopword. Theoretical Computer Science Ambiguous Natural Language Processing Word Sense Disambiguation Simplified Lesk Figures Figure 1 Figure 2 Figure 3 1. Introduction The presence of ambiguous words in the Indonesian language can make the meaning of a sentence unclear or confusing. This poses a challenge, as human perception has the linguistic ability to easily determine the correct meaning of ambiguous words based on context. Word Sense Disambiguation (WSD) is a subfield of Natural Language Processing (NLP) that aims to computationally identify the exact meaning of words based on their context [1]. A knowledge-based approach to WSD includes methods like the Lesk algorithm, which is widely used for its simple mechanism and ability to effectively handle polysemous words. There are two well-known variations of the Lesk algorithm: Original Lesk and Simplified Lesk . The Simplified Lesk algorithm is preferred due to its faster execution, lower computational time complexity, and higher accuracy in distinguishing word meanings compared to the Original Lesk algorithm [2]. Therefore, this study applies the Simplified Lesk algorithm to the task of WSD in Indonesian sentences. 2. Literature Study This chapter provides a literature review that forms the foundation of this research, covering Word Sense Disambiguation (WSD) , the Simplified Lesk Algorithm , and Evaluation Metrics . 2.1. Word Sense Disambiguation Lexical ambiguity occurs when a single word has multiple meanings. Resolving this ambiguity, known as Word Sense Disambiguation (WSD) , can improve various fields such as Machine Translation , Parsing , Natural Language Understanding (NLU) , and lexicography [3]. WSD is a subfield of NLP in which computer systems are designed to identify the correct meaning of a word in a specific context [4]. The WSD problem involves two main challenges: first, representing the different meanings of a word so an algorithm can interpret them, and second, determining which meaning fits the word's context [5]. 2.2. Preprocessing The preprocessing phase begins with tokenization , the process of breaking text into meaningful elements like words, phrases, and symbols [7]. This is followed by stopword removal , which eliminates irrelevant words, and stemming , which normalizes words to their base form [6]. The implementation of stemming and stopword removal can significantly influence the performance of a WSD system [8]. Stemming ensures that morphologically different forms of a word are treated as the same content word, while stopword removal prevents the content word matrix from becoming too sparse and eliminates unimportant words. 2.3. Simplified Lesk Algorithm The Lesk algorithm, published in 1986, is one of the earliest and most popular algorithms that use a machine-readable dictionary. The algorithm works by measuring the overlap between word sense definitions in a given context [9]. The Simplified Lesk algorithm is a variation of this approach. It runs a separate WSD process for each ambiguous word in the input text. The correct sense for each word is determined by finding the sense with the greatest overlap between its dictionary definition and the current context [10]. The process flow of the Simplified Lesk algorithm is illustrated in Fig. 1 . Before the algorithm is executed, sentences are preprocessed to generate a context. The input ambiguous word is then used to look up its definitions in a dictionary. The dictionary definitions (glosses) for all the possible meanings are examined, and the algorithm calculates the number of common words shared between the context and the gloss of each meaning [11]. The meaning with the highest number of overlapping words is considered the most appropriate sense [12]. 2.4. Evaluation Metric The standard evaluation method for WSD algorithms is the "exact match" criterion [13]. This method calculates overall accuracy by dividing the number of correctly predicted sense instances by the total number of instances [14]. A correct prediction is one where the predicted sense matches the true sense. The formula for calculating accuracy is: Based on Eq. (1), the "Number of Correct Predictions" refers to the instances where the algorithm correctly predicts the word sense, and the "Total Number of Instances" represents the total number of instances or examples used for evaluation. The higher the accuracy, the better the performance of the WSD algorithm in correctly identifying the appropriate word sense in the given context. 3. Methodology 3.1. Data Collection This study uses 300 secondary data points. This dataset includes 140 sentences with ambiguous Indonesian words from a scientific publication's test data. An additional 160 sentences were collected from various official articles, with the ambiguous words chosen according to the required criteria. The Actual Sense data was determined independently by human perception, selecting the most appropriate meaning from a dictionary that corresponded to the sentence's context. 3.2. System Architecture The overall architecture of the WSD system is shown in Fig. 2 . The process of building the system with the Simplified Lesk algorithm goes through several stages. First, the preprocessing stage prepares the data. The ambiguous words in the sentences are identified as target words. Then, their multiple meanings are retrieved from a dictionary to find the most appropriate meaning based on the sentence's context. Each word's meaning is processed by calculating the overlap between the meaning and the context. The system selects the meaning with the highest overlap as the predicted meaning that matches the ambiguous word. 4. Result and Discussion The testing involved four scenarios using 300 data points consisting of Indonesian sentences, ambiguous words, and the corresponding Actual Sense to generate context for determining the appropriate meaning. The WSD process was tested with the following preprocessing approaches: (1) stemming, (2) stopword removal, (3) both stemming and stopword removal, and (4) no preprocessing, i.e., without stemming or stopword removal. Based on the experimental configuration results, an analysis was conducted to calculate accuracy using the evaluation metrics. Each scenario was evaluated to determine the effectiveness and impact of the preprocessing steps on the WSD results. By comparing the outcomes, the study aimed to identify the most suitable preprocessing approach that yields the best WSD performance for the dataset. Figure 3 presents a comparison of the accuracy results obtained from each scenario, serving as the basis for selecting the best process for WSD using the Simplified Lesk algorithm in Indonesian sentences.. Based on the accuracy comparison shown in Fig. 3 , the best process for generating context and determining the correct sense is preprocessing with stemming and stopword removal, which achieved the highest accuracy of 58% across all tested data. Therefore, this experimental configuration is identified as the most effective approach for implementing WSD using the Simplified Lesk algorithm. Table 1 presents five sample results of WSD using the Simplified Lesk algorithm, along with explanations indicating whether the predicted meanings match the actual meanings. An additional analysis is then conducted for cases where the predicted meanings differ from the actual meanings.Table 1 . Sample Result of WSD Using Simplified Lesk NO WORD SENTENCE ACTUAL SENSE PREDICTED SENSE EXPLANA- TION 1 tahu Paman membawa banyak oleh- oleh makanan, yang paling terkenal adalah tahu sumedang makanan dari kedelai putih yang digiling halus-halus, direbus, dan dicetak kenal (akan); mengenal Not Match 2 halaman Halaman yang indah menjadi idaman setiap orang yang punya rumah pekarangan rumah (sekolah dan sebagainya); tanah di sekitar rumah (sekolah dan sebagainya) pekarangan rumah (sekolah dan sebagainya); tanah di sekitar rumah (sekolah dan sebagainya) Match 3 genting Saat kau ingin rumahmu terang alami, pakailah genting kaca di tutup atap rumah yang terbuat dari tanah liat yang dicetak dan dibakar, tutup atap rumah yang terbuat dari tanah liat yang dicetak dan dibakar, Match bagian tertentu atap rumahmu bermacam- macam bentuknya bermacam- macam bentuknya 4 bunga Setiap tahun, ani menerima bunga 5 persen dari bank imbalan jasa untuk penggunaan uang atau modal yang dibayar pada waktu tertentu berdasarkan ketentuan atau kesepakatan, umumnya dinyatakan sebagai persentase dari modal pokok bagian tumbuhan yang akan menjadi buah, biasanya elok warnanya dan harum baunya; kembang Not Match 5 babak Risqi Aris babak belur dihajar oleh tetangganya lecet (tentang kulit) bagian besar dalam suatu drama atau lakon (terdiri atas beberapa adegan) Not Match Based on the results in Table 1 , two main factors may have contributed to the incorrect predictions of the actual meanings. First, the lack of overlapping meanings with the context could be due to incomplete context generation and limited dictionary data, reducing the number of overlaps between the meanings and the context. Second, the selection of a meaning among candidates with the same level of overlap is not always accurate, as it tends to prioritize the first encountered meaning, which may not be the most appropriate for the context. To improve the accuracy of the WSD system using the Simplified Lesk algorithm, efforts should focus on expanding the dataset, enhancing the dictionary with more comprehensive word meanings, and optimizing the selection process to better identify the most suitable meaning. Additionally, integrating more sophisticated algorithms or incorporating advanced machine learning techniques could further enhance the system’s overall performance. 5. Conclusion The disambiguation of word meanings using the Simplified Lesk algorithm achieved an accuracy of 58%, which is relatively low and largely influenced by the quality of the data. The limited amount of contextual information and the lack of comprehensive word meanings in the dictionary contributed to this reduced accuracy, making it difficult for the system to identify the appropriate meanings effectively. To improve the accuracy of the WSD system, it is essential to develop a more extensive and comprehensive dataset that provides richer contextual information and includes complete word meanings in the dictionary, ensuring better identification of ambiguous words. References S. Basuki, A. Sofyan Kholimi, A. Eko Minarno, F. Sumadi, and M. Effendy, Word Sense Disambiguation (WSD) for Indonesian Homograph Word Meaning Determination by LESK Algorithm Application . 2019. A. Aliwy and A. Abbas, “Improvement WSD Dictionary Using Annotated Corpus and Testing it with Simplified Lesk Algorithm,” Academy and Industry Research Collaboration Center (AIRCC), Feb. 2015, pp. 89–97. doi: 10.5121/csit.2015.50409. A. Fujii and H. Tanaka, “Corpus-Based Word Sense Disambiguation,” 1998. D. Mccarthy, “Word sense disambiguation: An overview,” Linguistics and Language Compass , vol. 3, no. 2. Blackwell Publishing Inc., pp. 537–558, 2009. doi: 10.1111/j.1749-818X.2009.00131.x. A. Kilgarriff, “Gold standard datasets for evaluating word sense disambiguation programs,” 1998. A. Abdo, “Enhanced Word Sense Disambiguation Algorithm for Afaan Oromoo,” International Journal of Information Engineering and Electronic Business , vol. 15, no. 1, pp. 41–50, Feb. 2023, doi: 10.5815/ijieeb.2023.01.04. S. Kannan, V. Gurusamy, S. Vijayarani, J. Ilamathi, and M. Nithya, “Preprocessing Techniques for Text Mining,” 2014. R. Mahendra, H. Septiantri, H. A. Wibowo, R. Manurung, and M. Adriani, “Cross- Lingual and Supervised Learning Approach for Indonesian Word Sense Disambiguation Task,” 2018. D. J. Craggs, “An analysis and comparison of predominant word sense disambiguation algorithms,” 2011. [Online]. Available: https://ro.ecu.edu.au/theses_hons/4 Rada Mihalcea, “5 Knowledge-Based Methods for WSD,” 2007. A. Pal and D. Saha, An Approach to Automatic Text Summarization using WordNet . 2014. A. R. Pal, P. K. Maiti, and D. Saha, “An Approach to Automatic Text Summarization Using Simplified Lesk Algorithm and Wordnet,” International Journal of Control Theory and Computer Modeling , vol. 3, no. 5, pp. 15–23, Sep. 2013, doi: 10.5121/ijctcm.2013.3502. P. Resnik and D. Yarowsky, “A Perspective on Word Sense Disambiguation Methods and Their Evaluation,” 1997. S. Torres, A. Gelbukh, U. Profesional Adolfo-López Mateos, A. Juan de Dios Bátiz, and M. Othón de Mendizábal, “Comparing Similarity Measures for Original WSD Lesk Algorithm,” 2009. [Online]. Available: www.gelbukh.com Additional Declarations The authors declare no competing interests. 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-7472904","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":506449863,"identity":"0b691a45-d215-4188-99c1-d9209b6670fd","order_by":0,"name":"Nurul Akhni","email":"","orcid":"","institution":"Sriwijaya University","correspondingAuthor":false,"prefix":"","firstName":"Nurul","middleName":"","lastName":"Akhni","suffix":""},{"id":506450188,"identity":"9b221b20-b0f1-433c-835e-18732b7f2745","order_by":1,"name":"Abdiansah","email":"","orcid":"","institution":"Sriwijaya University","correspondingAuthor":false,"prefix":"","firstName":"","middleName":"","lastName":"Abdiansah","suffix":""},{"id":506450189,"identity":"8915dca1-fb01-4498-bbd4-d6487349fe41","order_by":2,"name":"Danny Matthew Saputra","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA9klEQVRIiWNgGAWjYDCCAyi8CiBmZm4gVgszEJ8B0YykaGFsAzEIaOE7fvbYYx6GOnlzifxjH37Oq43mbwdq+VGxDacWyTN56cY8DIcNd85IZp7Zu+147ozDjA2MPWdu49RicCDHTJqH4QDjhhvJzAy8247lNgC1MDO24dFy/g1IS509SAvj3znHcucT1HIDbAtzIkgLM29DTe4GQlokb7wxk5xjcDh5Z89jY2aZYwdyNwK1HMTnF77zOWYSbyrqbLezJz5mfFNTlzvv/OGDD35U4NYCAkw8BkAXQtiHweQBvOqBgPEHA1xLHSHFo2AUjIJRMAIBAELPWi2nBW0CAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0009-0000-1144-7092","institution":"Sriwijaya University","correspondingAuthor":true,"prefix":"","firstName":"Danny","middleName":"Matthew","lastName":"Saputra","suffix":""}],"badges":[],"createdAt":"2025-08-27 15:04:53","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-7472904/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7472904/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":90078652,"identity":"63f5b34d-fd6c-40d6-ba4f-6e7a64112a68","added_by":"auto","created_at":"2025-08-28 08:27:38","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":133418,"visible":true,"origin":"","legend":"\u003cp\u003eSimplified Lesk Process (Aliwy \u0026amp; Abbas, 2015)\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7472904/v1/8b9eaba9b4ae0a3ea6320741.png"},{"id":90078651,"identity":"acc84365-b571-4517-8550-b19f465df670","added_by":"auto","created_at":"2025-08-28 08:27:38","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":53923,"visible":true,"origin":"","legend":"\u003cp\u003eWSD Architecture System\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7472904/v1/4e698abdc70713cd18d95c95.png"},{"id":90078819,"identity":"b7671144-d504-4107-820d-32ca6c9c7edc","added_by":"auto","created_at":"2025-08-28 08:35:38","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":67612,"visible":true,"origin":"","legend":"\u003cp\u003eScenario Result Comparison\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7472904/v1/5a36e7c11df41f9994870ea9.png"},{"id":90079743,"identity":"843affa2-1551-48b3-bec2-74a3bf5aef48","added_by":"auto","created_at":"2025-08-28 08:43:43","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":808163,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7472904/v1/c62a09ec-3041-4a22-9efc-117eaf04dfa2.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cem\u003eWord Sense Disambiguation \u003c/em\u003e(WSD) in Indonesian Sentences Using Simplified Lesk Algorithm\u003c/p\u003e","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eThe presence of ambiguous words in the Indonesian language can make the meaning of a sentence unclear or confusing. This poses a challenge, as human perception has the linguistic ability to easily determine the correct meaning of ambiguous words based on context. \u003cb\u003eWord Sense Disambiguation (WSD)\u003c/b\u003e is a subfield of Natural Language Processing (NLP) that aims to computationally identify the exact meaning of words based on their context [1].\u003c/p\u003e\u003cp\u003eA knowledge-based approach to WSD includes methods like the Lesk algorithm, which is widely used for its simple mechanism and ability to effectively handle polysemous words. There are two well-known variations of the Lesk algorithm: \u003cb\u003eOriginal Lesk\u003c/b\u003e and \u003cb\u003eSimplified Lesk\u003c/b\u003e. The Simplified Lesk algorithm is preferred due to its faster execution, lower computational time complexity, and higher accuracy in distinguishing word meanings compared to the Original Lesk algorithm [2]. Therefore, this study applies the Simplified Lesk algorithm to the task of WSD in Indonesian sentences.\u003c/p\u003e"},{"header":"2. Literature Study","content":"\u003cp\u003eThis chapter provides a literature review that forms the foundation of this research, covering \u003cb\u003eWord Sense Disambiguation (WSD)\u003c/b\u003e, the \u003cb\u003eSimplified Lesk Algorithm\u003c/b\u003e, and \u003cb\u003eEvaluation Metrics\u003c/b\u003e.\u003c/p\u003e\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1. Word Sense Disambiguation\u003c/h2\u003e\u003cp\u003e\u003cb\u003eLexical ambiguity\u003c/b\u003e occurs when a single word has multiple meanings. Resolving this ambiguity, known as \u003cb\u003eWord Sense Disambiguation (WSD)\u003c/b\u003e, can improve various fields such as \u003cb\u003eMachine Translation\u003c/b\u003e, \u003cb\u003eParsing\u003c/b\u003e, \u003cb\u003eNatural Language Understanding (NLU)\u003c/b\u003e, and \u003cb\u003elexicography\u003c/b\u003e [3]. WSD is a subfield of NLP in which computer systems are designed to identify the correct meaning of a word in a specific context [4]. The WSD problem involves two main challenges: first, representing the different meanings of a word so an algorithm can interpret them, and second, determining which meaning fits the word's context [5].\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2. Preprocessing\u003c/h2\u003e\u003cp\u003eThe preprocessing phase begins with \u003cb\u003etokenization\u003c/b\u003e, the process of breaking text into meaningful elements like words, phrases, and symbols [7]. This is followed by \u003cb\u003estopword removal\u003c/b\u003e, which eliminates irrelevant words, and \u003cb\u003estemming\u003c/b\u003e, which normalizes words to their base form [6]. The implementation of stemming and stopword removal can significantly influence the performance of a WSD system [8]. Stemming ensures that morphologically different forms of a word are treated as the same content word, while stopword removal prevents the content word matrix from becoming too sparse and eliminates unimportant words.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e2.3. Simplified Lesk Algorithm\u003c/h2\u003e\u003cp\u003eThe Lesk algorithm, published in 1986, is one of the earliest and most popular algorithms that use a machine-readable dictionary. The algorithm works by measuring the overlap between word sense definitions in a given context [9]. The \u003cb\u003eSimplified Lesk algorithm\u003c/b\u003e is a variation of this approach. It runs a separate WSD process for each ambiguous word in the input text. The correct sense for each word is determined by finding the sense with the greatest overlap between its dictionary definition and the current context [10]. The process flow of the Simplified Lesk algorithm is illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eBefore the algorithm is executed, sentences are preprocessed to generate a context. The input ambiguous word is then used to look up its definitions in a dictionary. The dictionary definitions (glosses) for all the possible meanings are examined, and the algorithm calculates the number of common words shared between the context and the gloss of each meaning [11]. The meaning with the highest number of overlapping words is considered the most appropriate sense [12].\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e2.4. Evaluation Metric\u003c/h2\u003e\u003cp\u003eThe standard evaluation method for WSD algorithms is the \"exact match\" criterion [13]. This method calculates overall accuracy by dividing the number of correctly predicted sense instances by the total number of instances [14]. A correct prediction is one where the predicted sense matches the true sense. The formula for calculating accuracy is:\u003c/p\u003e\u003cp\u003e\u003cimg src=\"data:image/png;base64,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\" width=\"480\" height=\"37\"\u003e\u003c/p\u003e\u003cp\u003eBased on Eq.\u0026nbsp;(1), the \"Number of Correct Predictions\" refers to the instances where the algorithm correctly predicts the word sense, and the \"Total Number of Instances\" represents the total number of instances or examples used for evaluation. The higher the accuracy, the better the performance of the WSD algorithm in correctly identifying the appropriate word sense in the given context.\u003c/p\u003e\u003c/div\u003e"},{"header":"3. Methodology","content":"\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003e3.1. Data Collection\u003c/h2\u003e\u003cp\u003eThis study uses 300 secondary data points. This dataset includes 140 sentences with ambiguous Indonesian words from a scientific publication's test data. An additional 160 sentences were collected from various official articles, with the ambiguous words chosen according to the required criteria. The \u003cb\u003eActual Sense\u003c/b\u003e data was determined independently by human perception, selecting the most appropriate meaning from a dictionary that corresponded to the sentence's context.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e3.2. System Architecture\u003c/h2\u003e\u003cp\u003eThe overall architecture of the WSD system is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. The process of building the system with the Simplified Lesk algorithm goes through several stages.\u003c/p\u003e\u003cp\u003eFirst, the \u003cb\u003epreprocessing\u003c/b\u003e stage prepares the data. The ambiguous words in the sentences are identified as target words. Then, their multiple meanings are retrieved from a dictionary to find the most appropriate meaning based on the sentence's context. Each word's meaning is processed by calculating the overlap between the meaning and the context. The system selects the meaning with the highest overlap as the predicted meaning that matches the ambiguous word.\u003c/p\u003e\u003c/div\u003e"},{"header":"4. Result and Discussion","content":"\u003cp\u003eThe testing involved four scenarios using 300 data points consisting of Indonesian sentences, ambiguous words, and the corresponding Actual Sense to generate context for determining the appropriate meaning. The WSD process was tested with the following preprocessing approaches: (1) stemming, (2) stopword removal, (3) both stemming and stopword removal, and (4) no preprocessing, i.e., without stemming or stopword removal. Based on the experimental configuration results, an analysis was conducted to calculate accuracy using the evaluation metrics.\u003c/p\u003e\u003cp\u003eEach scenario was evaluated to determine the effectiveness and impact of the preprocessing steps on the WSD results. By comparing the outcomes, the study aimed to identify the most suitable preprocessing approach that yields the best WSD performance for the dataset. Figure\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e presents a comparison of the accuracy results obtained from each scenario, serving as the basis for selecting the best process for WSD using the Simplified Lesk algorithm in Indonesian sentences..\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eBased on the accuracy comparison shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, the best process for generating context and determining the correct sense is preprocessing with stemming and stopword removal, which achieved the highest accuracy of 58% across all tested data. Therefore, this experimental configuration is identified as the most effective approach for implementing WSD using the Simplified Lesk algorithm.\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\u003epresents five sample results of WSD using the Simplified Lesk algorithm, along with explanations indicating whether the predicted meanings match the actual meanings. An additional analysis is then conducted for cases where the predicted meanings differ from the actual meanings.Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Sample Result of WSD Using Simplified Lesk\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\u003eNO\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eWORD\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSENTENCE\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cem\u003eACTUAL SENSE\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003ePREDICTED\u003c/em\u003e\u003c/p\u003e\u003cp\u003e\u003cem\u003eSENSE\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eEXPLANA-\u003c/p\u003e\u003cp\u003eTION\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\u003etahu\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003ePaman membawa banyak oleh- oleh makanan, yang paling terkenal adalah tahu sumedang\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003emakanan dari kedelai putih yang digiling halus-halus, direbus, dan dicetak\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003ekenal (akan); mengenal\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eNot Match\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\u003ehalaman\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eHalaman yang indah menjadi idaman setiap orang yang punya rumah\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003epekarangan rumah (sekolah dan sebagainya); tanah di sekitar rumah (sekolah dan sebagainya)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003epekarangan rumah (sekolah dan sebagainya); tanah di sekitar rumah (sekolah dan sebagainya)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eMatch\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\u003egenting\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSaat kau ingin rumahmu terang alami, pakailah\u003c/p\u003e\u003cp\u003egenting kaca di\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003etutup atap rumah yang terbuat dari tanah liat yang dicetak dan\u003c/p\u003e\u003cp\u003edibakar,\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003etutup atap rumah yang terbuat dari tanah liat yang dicetak dan\u003c/p\u003e\u003cp\u003edibakar,\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eMatch\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\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"No\" id=\"Tabb\" 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\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003ebagian tertentu atap rumahmu\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003ebermacam- macam bentuknya\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003ebermacam- macam bentuknya\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003ebunga\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSetiap tahun, ani menerima bunga 5 persen dari bank\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eimbalan jasa untuk penggunaan uang atau modal yang dibayar pada waktu tertentu berdasarkan ketentuan atau kesepakatan, umumnya dinyatakan sebagai persentase dari modal pokok\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003ebagian tumbuhan yang akan menjadi buah, biasanya elok warnanya dan harum baunya; kembang\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eNot Match\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\u003ebabak\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eRisqi Aris babak belur dihajar oleh tetangganya\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003elecet (tentang kulit)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003ebagian besar dalam suatu drama atau lakon (terdiri atas beberapa adegan)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eNot Match\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\u003eBased on the results in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, two main factors may have contributed to the incorrect predictions of the actual meanings. First, the lack of overlapping meanings with the context could be due to incomplete context generation and limited dictionary data, reducing the number of overlaps between the meanings and the context. Second, the selection of a meaning among candidates with the same level of overlap is not always accurate, as it tends to prioritize the first encountered meaning, which may not be the most appropriate for the context. To improve the accuracy of the WSD system using the Simplified Lesk algorithm, efforts should focus on expanding the dataset, enhancing the dictionary with more comprehensive word meanings, and optimizing the selection process to better identify the most suitable meaning. Additionally, integrating more sophisticated algorithms or incorporating advanced machine learning techniques could further enhance the system\u0026rsquo;s overall performance.\u003c/p\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eThe disambiguation of word meanings using the Simplified Lesk algorithm achieved an accuracy of 58%, which is relatively low and largely influenced by the quality of the data. The limited amount of contextual information and the lack of comprehensive word meanings in the dictionary contributed to this reduced accuracy, making it difficult for the system to identify the appropriate meanings effectively. To improve the accuracy of the WSD system, it is essential to develop a more extensive and comprehensive dataset that provides richer contextual information and includes complete word meanings in the dictionary, ensuring better identification of ambiguous words.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eS. Basuki, A. Sofyan Kholimi, A. Eko Minarno, F. Sumadi, and M. Effendy, \u003cem\u003eWord Sense Disambiguation (WSD) for Indonesian Homograph Word Meaning Determination by LESK Algorithm Application\u003c/em\u003e. 2019.\u003c/li\u003e\n \u003cli\u003eA. Aliwy and A. Abbas, \u0026ldquo;Improvement WSD Dictionary Using Annotated Corpus and Testing it with Simplified Lesk Algorithm,\u0026rdquo; Academy and Industry Research Collaboration Center (AIRCC), Feb. 2015, pp. 89\u0026ndash;97. doi: 10.5121/csit.2015.50409.\u003c/li\u003e\n \u003cli\u003eA. Fujii and H. Tanaka, \u0026ldquo;Corpus-Based Word Sense Disambiguation,\u0026rdquo; 1998.\u003c/li\u003e\n \u003cli\u003eD. Mccarthy, \u0026ldquo;Word sense disambiguation: An overview,\u0026rdquo; \u003cem\u003eLinguistics and Language Compass\u003c/em\u003e, vol. 3, no. 2. Blackwell Publishing Inc., pp. 537\u0026ndash;558, 2009. doi: 10.1111/j.1749-818X.2009.00131.x.\u003c/li\u003e\n \u003cli\u003eA. Kilgarriff, \u0026ldquo;Gold standard datasets for evaluating word sense disambiguation programs,\u0026rdquo; 1998.\u003c/li\u003e\n \u003cli\u003eA. Abdo, \u0026ldquo;Enhanced Word Sense Disambiguation Algorithm for Afaan Oromoo,\u0026rdquo; \u003cem\u003eInternational Journal of Information Engineering and Electronic Business\u003c/em\u003e, vol. 15, no. 1, pp. 41\u0026ndash;50, Feb. 2023, doi: 10.5815/ijieeb.2023.01.04.\u003c/li\u003e\n \u003cli\u003eS. Kannan, V. Gurusamy, S. Vijayarani, J. Ilamathi, and M. Nithya, \u0026ldquo;Preprocessing Techniques for Text Mining,\u0026rdquo; 2014.\u003c/li\u003e\n \u003cli\u003eR. Mahendra, H. Septiantri, H. A. Wibowo, R. Manurung, and M. Adriani, \u0026ldquo;Cross- Lingual and Supervised Learning Approach for Indonesian Word Sense Disambiguation Task,\u0026rdquo; 2018.\u003c/li\u003e\n \u003cli\u003eD. J. Craggs, \u0026ldquo;An analysis and comparison of predominant word sense disambiguation algorithms,\u0026rdquo; 2011. [Online]. Available: https://ro.ecu.edu.au/theses_hons/4\u003c/li\u003e\n \u003cli\u003eRada Mihalcea, \u0026ldquo;5 Knowledge-Based Methods for WSD,\u0026rdquo; 2007.\u003c/li\u003e\n \u003cli\u003eA. Pal and D. Saha, \u003cem\u003eAn Approach to Automatic Text Summarization using WordNet\u003c/em\u003e. 2014.\u003c/li\u003e\n \u003cli\u003eA. R. Pal, P. K. Maiti, and D. Saha, \u0026ldquo;An Approach to Automatic Text Summarization Using Simplified Lesk Algorithm and Wordnet,\u0026rdquo; \u003cem\u003eInternational Journal of Control Theory and Computer Modeling\u003c/em\u003e, vol. 3, no. 5, pp. 15\u0026ndash;23, Sep. 2013, doi: 10.5121/ijctcm.2013.3502.\u003c/li\u003e\n \u003cli\u003eP. Resnik and D. Yarowsky, \u0026ldquo;A Perspective on Word Sense Disambiguation Methods and Their Evaluation,\u0026rdquo; 1997.\u003c/li\u003e\n \u003cli\u003eS. Torres, A. Gelbukh, U. Profesional Adolfo-L\u0026oacute;pez Mateos, A. Juan de Dios B\u0026aacute;tiz, and M. Oth\u0026oacute;n de Mendiz\u0026aacute;bal, \u0026ldquo;Comparing Similarity Measures for Original WSD Lesk Algorithm,\u0026rdquo; 2009. [Online]. Available: www.gelbukh.com\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Sriwijaya University","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":"Ambiguous, Natural Language Processing, Word Sense Disambiguation, Simplified Lesk","lastPublishedDoi":"10.21203/rs.3.rs-7472904/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7472904/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe Indonesian language contains several words with inherent ambiguity, meaning they possess more than one possible interpretation. Word Sense Disambiguation (WSD), a branch of Natural Language Processing (NLP), deals with the challenge of resolving this ambiguity and identifying the precise meaning of a word based on its context. Among the algorithms used for WSD, the Simplified Lesk algorithm stands out as particularly popular. To assess its effectiveness, tests were conducted using the Kamus Besar Bahasa Indonesia (KBBI) as a reference for word definitions, and a dataset of 300 Indonesian sentences containing ambiguous words and their respective meanings as determined by human perception. The research reveals that the configuration of the preprocessing phase plays a crucial role in accurately identifying the intended meaning. After evaluation, the overall accuracy achieved was 58% for the dataset, incorporating preprocessing techniques such as stemming and stopword.\u003c/p\u003e","manuscriptTitle":"Word Sense Disambiguation (WSD) in Indonesian Sentences Using Simplified Lesk Algorithm","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-28 08:27:34","doi":"10.21203/rs.3.rs-7472904/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":"9c4a74c9-725e-46f6-a744-c3c3e791355f","owner":[],"postedDate":"August 28th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":53795314,"name":"Theoretical Computer Science"}],"tags":[],"updatedAt":"2025-08-28T08:27:34+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-28 08:27:34","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7472904","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7472904","identity":"rs-7472904","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}