DEPRESSION DETECTION ON TWITTER USING GATED RECURRENT UNIT

Alfransis Perugia Bennybeng Holle; Warih Maharani

doi:10.52436/1.jutif.2024.5.1.1187

Alfransis Perugia Bennybeng Holle Informatics, Informatics Faculty, Universitas Telkom, Indonesia, Indonesia
Warih Maharani Informatics, Informatics Faculty, Universitas Telkom, Indonesia, Indonesia

DOI: https://doi.org/10.52436/1.jutif.2024.5.1.1187

Keywords: Dataset, Depression, Evaluation, GRU, Twitter

Abstract

In the present era, technological advancements have significantly impacted society, particularly in the use of social media. One popular social media platform is Twitter, where people could share moments, thoughts, and statuses. However, since the COVID-19 pandemic, the usage of Twitter increased, and some users began exhibiting symptoms of depression. The condition of depression required a means to channel emotions that could assist users in coping. By employing the GRU method and Word2Vec feature extraction, we developed a depression detection system capable of analyzing users' Twitter posts and identifying potential signs of depression. The dataset used in this research was obtained from 165 participants who agreed to utilize their personal Twitter data and completed a questionnaire based on the Depression Anxiety and Stress Scales-42 (DASS-42). The questionnaire results served as labels that were processed for Word2Vec feature extraction and subsequently fed into the GRU model. The evaluation revealed an accuracy rate of 57.58% and an f1-score of 56.25. By using the bidirectional layer in the model, there is an improvement in precision, recall, and f1-score values.

Downloads

Download data is not yet available.

References

J. Sinuhaji, “Dunia Terisolasi Pandemi Covid-19, Pengguna Twitter Meningkat,” PikiranRakyat.com. https://www.pikiran-rakyat.com/teknologi/pr-01634954/dunia-terisolasi-pandemi-covid-19-pengguna-Twitter-meningkat (accessed Dec. 02, 2022).

M. A. Rizaty, “Pengguna Twitter di Indonesia Capai 18,45 Juta pada 2022,” DataIndonesia.id. https://dataindonesia.id/digital/detail/pengguna-twitter-di-indonesia-capai-1845-juta-pada-2022 (accessed Dec. 06, 2022).

P. Ananda, “WHO Sebut Pandemi Covid-19 Sebabkan Tingkat Depresi Naik 25%.” https://www.okezone.com/tren/read/2022/03/15/620/2561866/who-sebut-pandemi-covid-19-sebabkan-tingkat-depresi-naik-25 (accessed Dec. 02, 2022).

M. A. Rizaty, “Survei: 1 dari 3 Remaja Indonesia Punya Masalah Kesehatan Mental Artikel ini telah tayang di DSurvei: 1 dari 3 Remaja Indonesia Punya Masalah Kesehatan Mental,” dataindonesia.id. https://dataindonesia.id/ragam/detail/survei-1-dari-3-remaja-indonesia-punya-masalah-kesehatan-mental (accessed Dec. 05, 2022).

A. Pragholapati, “Covid-19 impact on students,” 2020. doi: 10.35542/osf.io/895ed.

K. Dianovinina and F. Psikologi, “Depresi pada Remaja: Gejala dan Permasalahannya Depression in Adolescent: Symptoms and the Problems,” 2018.

S. Yang, X. Yu, and Y. Zhou, “LSTM and GRU Neural Network Performance Comparison Study: Taking Yelp Review Dataset as an Example,” in Proceedings - 2020 International Workshop on Electronic Communication and Artificial Intelligence, IWECAI 2020, Institute of Electrical and Electronics Engineers Inc., Jun. 2020, pp. 98–101. doi: 10.1109/IWECAI50956.2020.00027.

“Mengenali Depresi dari Cuitan Seseorang di Twitter,” nationalgeorgraphic.grid.id. https://nationalgeographic.grid.id/read/13309312/mengenali-depresi-dari-cuitan-seseorang-di-twitter (accessed Dec. 05, 2022).

J. Patihullah and E. Winarko, “Hate Speech Detection for Indonesia Tweets Using Word Embedding And Gated Recurrent Unit,” IJCCS (Indonesian Journal of Computing and Cybernetics Systems), vol. 13, no. 1, p. 43, Jan. 2019, doi: 10.22146/ijccs.40125.

G. Qorik, O. Pratamasunu, F. N. Fajri, D. Puji, and K. Sari, “Deteksi Tangan Otomatis Pada Video Percakapan Bahasa Isyarat Indonesia Menggunakan Metode Deep Gated Recurrent Unit (GRU),” 2022. [Online]. Available: https://jurnal.pcr.ac.id/index.php/jkt/

P. E. Sumolang and W. Maharani, “Depression detection on Twitter using Bidirectional long short term memory,” Building of Informatics, Technology and Science (BITS), vol. 4, no. 2, pp. 369–376, 2022. doi:10.47065/bits.v4i2.1850.

H. H. Windjatika and W. Maharani, “Depression detection on social media Twitter using long short-term memory,” JURNAL MEDIA INFORMATIKA BUDIDARMA, vol. 6, no. 4, p. 1835, 2022. doi:10.30865/mib.v6i4.4457.

A. A. Simanungkalit, W. Maharani, and P. H. Gani, “Depression Detection on Twitter Social Media Platform using Bidirectional Long-Short Term Memory,” JINAV: Journal of Information and Visualization, vol. 3, no. 2, pp. 190–203, Dec. 2022, doi: 10.35877/454ri.jinav1503.

A. Nurdin, B. Anggo, S. Aji, A. Bustamin, and Z. Abidin, “PERBANDINGAN KINERJA WORD EMBEDDING WORD2VEC, GLOVE, DAN FASTTEXT PADA KLASIFIKASI TEKS,” Jurnal TEKNOKOMPAK, vol. 14, no. 2, p. 74, 2020.

S. Kusumadewi and H. Wahyuningsih, “Model Sistem Pendukung Keputusan Kelompok untuk Penilaian Gangguan Depresii, Kecemasan dan Stress Berdasarkan DASS-42,” Jurnal Teknologi Informasi dan Ilmu Komputer, vol. 7, no. 2, p. 219, 2020, doi: 10.25126/jtiik.2020721052.

D. Jatnika, M. A. Bijaksana, and A. A. Suryani, “Word2vec model analysis for semantic similarities in English words,” in Procedia Computer Science, Elsevier B.V., 2019, pp. 160–167. doi: 10.1016/j.procs.2019.08.153.

T. Mikolov, I. Sutskever, K. Chen, G. Corrado, and J. Dean, “Distributed Representations of Words and Phrases and their Compositionality,” Oct. 2013, [Online]. Available: http://arxiv.org/abs/1310.4546

J. Chen, H. Jing, Y. Chang, and Q. Liu, “Gated recurrent unit based recurrent neural network for remaining useful life prediction of nonlinear deterioration process,” Reliab Eng Syst Saf, vol. 185, pp. 372–382, May 2019, doi: 10.1016/j.ress.2019.01.006.

M. Hasnain, M. F. Pasha, I. Ghani, M. Imran, M. Y. Alzahrani, and R. Budiarto, “Evaluating Trust Prediction and Confusion Matrix Measures for Web Services Ranking,” IEEE Access, vol. 8, pp. 90847–90861, 2020, doi: 10.1109/ACCESS.2020.2994222.

Q. Gu, L. Zhu, and Z. Cai, “Evaluation Measures of the Classification Performance of Imbalanced Data Sets,” 2009.