{ "article_id": 1461, "public_id": "e80cd809-5942-4369-bc05-eeb05137c2bc", "internal_identifier": null, "source_identifier": "90:10", "status": "published", "type": "research_article", "title": "ELECTRA ve XLNET Modellerini Kullanarak X Verilerinden İntihara Meyilli İçerikleri Tespit Etme", "title_secondary": null, "abstract": "İntihar konusu, birçok farklı disiplin tarafından incelenen ve tarihin her döneminde karşılaşılan bir konu olmuştur. Teknolojinin gelişimi ve akıllı telefonların her yerde bulunması sebebiyle, kullanıcılar, özellikle çevrimiçi sosyal platformlar aracılığıyla duygu ve düşüncelerini rahatlıkla ifade edebilmişlerdir. Paylaşılan bu düşüncelerden biri de intihar düşüncesidir. İntiharın önlenebilmesi ve genç popülasyonda artan intihar oranlarının kontrol altına alınabilmesi için son zamanlarda sosyal medyada paylaşılan intihar düşüncelerinin tespiti üzerine yapılan araştırmaların sayısı her geçen gün artmaktadır. Bu çalışmada, en popüler sosyal medya platformlarından birisi olan X’te paylaşılan gönderilerden hazırlanan bir veri seti üzerinde makine öğrenmesi ve doğal dil işleme teknikleri kullanılarak bu paylaşımı yapan kişilerin intihara meyilli olup olmadığının tespitinin yapılmasına odaklanılmıştır. Önceden eğitilmiş transformatör yöntemlerinin klasik makine öğrenmesi yöntemleri ile karşılaştırmalı bir analizi sunulmuştur. Tüm modeller içerisinde en yüksek doğruluk oranı, ELECTRA modeli ile %96.61 olarak elde edilmiştir. Bu bulgular, intihara meyilli tweetlerin tespitinde transformatör modellerinin potansiyelinin yüksek olduğunu göstermekte ve özellikle ELECTRA'nın bu alanda etkili bir araç olabileceğini ortaya koymaktadır.", "language": "tr", "english_title": "Detecting Suicidal Content from X Data Using ELECTRA and XLNET Models", "english_abstract": "Suicide has been a subject studied by many different disciplines and encountered throughout history. Due to the development of technology and the ubiquity of smartphones, users have been able to easily express their feelings and thoughts, especially through online social platforms. One of these shared thoughts is suicidal ideation. In order to prevent suicide and control the increasing suicide rates in the young population, the number of studies on detecting suicidal thoughts shared on social media is increasing every day. In this study, the focus is on determining whether the individuals who made these posts on X, one of the most popular social media platforms, are suicidal using machine learning and natural language processing techniques on a dataset prepared from these posts. A comparative analysis of pre-trained transformer methods with classical machine learning methods is presented. Among all models, the highest accuracy rate was obtained with the ELECTRA model at 96.61%. These findings demonstrate the high potential of transformer models in detecting suicidal tweets, and particularly highlight that ELECTRA could be an effective tool in this area.", "doi": "10.36287/setsci.6.1.019", "doi_url": "https://doi.org/10.36287/setsci.6.1.019", "resource_url": "https://www.set-science.com/a/e80cd809-5942-4369-bc05-eeb05137c2bc", "pdf_url": "https://www.set-science.com/a/e80cd809-5942-4369-bc05-eeb05137c2bc/pdf", "og_image": "https://www.set-science.com/assets/img/og-card-v2.png", "authors": [ { "public_id": "00348531-7d39-4c57-9c20-44bd9a0d39d3", "given": "Tolga", "family": "Aydın", "full": "Tolga Aydın", "literal": null, "orcid": null, "orcid_url": null, "email": "atolga@atauni.edu.tr", "corresponding": false, "sequence": "first", "affiliations": [ { "name": "Atatürk Üniversitesi", "department": null, "country": null, "country_code": null, "ror": null, "ror_url": null } ] }, { "public_id": "5a178664-3a2a-44fb-9d2a-caa84b491921", "given": "Muhammet Coşkun", "family": "Irmak", "full": "Muhammet Coşkun Irmak", "literal": null, "orcid": null, "orcid_url": null, "email": "coskunirmak@yyu.edu.tr", "corresponding": true, "sequence": "additional", "affiliations": [ { "name": "Van Yüzüncü Yıl Üniversitesi", "department": null, "country": null, "country_code": null, "ror": null, "ror_url": null } ] } ], "keywords": [ "İntihar tespiti", "Doğal dil işleme", "ELECTRA", "XLNET", "Transformatör model" ], "translations": [ { "language": "en", "title": "Detecting Suicidal Content from X Data Using ELECTRA and XLNET Models", "abstract": "Suicide has been a subject studied by many different disciplines and encountered throughout history. Due to the development of technology and the ubiquity of smartphones, users have been able to easily express their feelings and thoughts, especially through online social platforms. One of these shared thoughts is suicidal ideation. In order to prevent suicide and control the increasing suicide rates in the young population, the number of studies on detecting suicidal thoughts shared on social media is increasing every day. In this study, the focus is on determining whether the individuals who made these posts on X, one of the most popular social media platforms, are suicidal using machine learning and natural language processing techniques on a dataset prepared from these posts. A comparative analysis of pre-trained transformer methods with classical machine learning methods is presented. Among all models, the highest accuracy rate was obtained with the ELECTRA model at 96.61%. These findings demonstrate the high potential of transformer models in detecting suicidal tweets, and particularly highlight that ELECTRA could be an effective tool in this area.", "keywords": [ "Suicide detection", "Natural language processing", "ELECTRA", "XLNET", "Transformer model" ] } ], "subjects": [], "subject_names": [], "published_date": "2023-12-29", "published_year": "2023", "received_date": null, "accepted_date": null, "volume": "15", "issue": null, "page_first": "45", "page_last": "50", "article_number": null, "conference": { "short_name": "BMYZ2023", "full_title": "Cognitive Models and Artificial Intelligence Conference", "name": "Cognitive Models and Artificial Intelligence Conference", "acronym": "BMYZ2023", "location": "Ankara, Türkiye", "start_date": "2023-10-26", "end_date": null, "public_id": "dc3a47c3-f4b3-4059-b610-f99aca33d7a0" }, "proceedings": { "title": "Cognitive Models and Artificial Intelligence Conference", "series_title": "SETSCI Conference Proceedings", "publisher": "SET Technology", "isbn": "978-605-69925-7-5", "eisbn": null, "issn": null, "eissn": "2687-5527", "doi": null, "doi_prefix": "10.36287", "slug": "setsci-volume-15-2023-legacy-90", "public_id": "33fc699d-0ab3-4057-b2e0-644dc89f900d", "url": "https://www.set-science.com/p/setsci-volume-15-2023-legacy-90" }, "license": { "name": "CC BY 4.0", "url": "https://creativecommons.org/licenses/by/4.0/", "code": "CC-BY-4.0", "open_access": true }, "copyright": { "holder": "SET Technology", "year": "2023" }, "funding_statement": null, "funders": [], "references": [ { "order": 1, "raw": "Rabani, S. T., Khan, Q. R., and Khanday, A. M. U. D. “Detection of suicidal ideation on Twitter using machine learning & ensemble approaches”. Baghdad science journal, 17(4), 1328-1328, 2020.\r\n\r\nHaque, R., Islam, N., Islam, M., and Ahsan, M. M. “A comparative analysis on suicidal ideation detection using NLP, machine, and deep learning.” Technologies, 10(3), 57,2022.\r\n\r\nBaghdadi, N. A., Malki, A., Balaha, H. M., AbdulAzeem, Y., Badawy, M., and Elhosseini, M. “An optimized deep learning approach for suicide detection through Arabic tweets.” PeerJ Computer Science, 8, e1070, 2022.\r\n\r\nAbdulsalam, A., Alhothali, A., and Al-Ghamdi, S. “Detecting Suicidality in Arabic Tweets Using Machine Learning and Deep Learning Techniques.” arXiv preprint arXiv:2309.00246, 2023.\r\n\r\nAbdulsalam, A., and Alhothali, A. “Suicidal ideation detection on social media: A review of machine learning methods.” arXiv preprint arXiv:2201.10515, 2022.\r\n\r\nRamírez-Cifuentes, D., Freire, A., Baeza-Yates, R., Puntí, J., Medina-Bravo, P., Velazquez, D. A., ... and Gonzàlez, J. “Detection of suicidal ideation on social media: multimodal, relational, and behavioral analysis.” Journal of medical internet research, 22(7), e17758, 2020.\r\n\r\nChatterjee, M., Kumar, P., Samanta, P., and Sarkar, D. “Suicide ideation detection from online social media: A multi-modal feature based technique.” International Journal of Information Management Data Insights, 2(2), 100103, 2022.\r\n\r\nSabri, N. M., and Mohamad, N. A. “Detection of Suicidal Tweets Based On Naïve Bayes Algorithm.” International Journal of Advanced Research in Technology and Innovation, 4(3), 47-59, 2022.\r\n\r\nDeepa, J., Shriraaman, S., Shruti, V. V., and Vasanth, G. “Detecting and Determining Degree of Suicidal Ideation on Tweets Using LSTM and Machine Learning Models.” Journal of Survey in Fisheries Sciences, 10(2S), 3217-3224, 2023.\r\n\r\nPriyamvada, B., Singhal, S., Nayyar, A., Jain, R., Goel, P., Rani, M., and Srivastava, M. “Stacked CNN-LSTM approach for prediction of suicidal ideation on social media.” Multimedia Tools and Applications, 1-22, 2023.\r\n\r\nMahmud, S. A., Suicidal Tweet Detection Dataset, (Son Erişim Tarihi: 30.09.2023), https://www.kaggle.com/datasets/aunanya875/suicidal-tweet-detection-dataset \r\n\r\nNaveen, S., Kiran, M. S. R., Indupriya, M., Manikanta, T. V., and Sudeep, P. V. “Transformer models for enhancing AttnGAN based text to image generation.” Image and Vision Computing, 115, 104284, 2021.\r\n\r\nClark, K., Luong, M. T., Le, Q. V., and Manning, C. D. “Electra: Pre-training text encoders as discriminators rather than generators.” arXiv preprint arXiv:2003.10555, 2020.\r\n\r\nLi, X., Li, L., Ma, W., & Wang, W. “Two-phase flow patterns identification in porous media using feature extraction and SVM.” International Journal of Multiphase Flow, 104222, 2022.\r\n\r\nSingh, N., Jena, S., and Panigrahi, C. K. “A novel application of Decision Tree classifier in solar irradiance prediction.” Materials Today: Proceedings, 58, 316-323, 2022.\r\n\r\nKavzoğlu, T., and Çölkesen, İ. “Karar ağaçları ile uydu görüntülerinin sınıflandırılması.” Harita Teknolojileri Elektronik Dergisi, 2(1), 36-45, 2010.\r\n\r\nYangın, G. “XGBoost ve Karar Ağacı Tabanlı Algoritmaların Diyabet Veri Setleri Üzerine Uygulaması” Mimar Sinan Güzel Sanatlar Üniversitesi, Fen Bilimleri Enstitüsü, İstatistik Anabilim Dalı, İstanbul, 2019.", "doi": null, "title": null, "authors": null, "source": null, "year": null, "volume": null, "pages": null } ], "statements": { "acknowledgements": null, "conflict_of_interest": null, "data_availability": null, "ai_disclosure": null }, "credit": [], "history": { "published": "2023-12-29" } }