Depression Early Warning System Based on Social Media Using Multinomial Naïve Bayes Algorithm

DOI

10.2991/978-94-6463-445-7_8How to use a DOI?

Keywords

Social media and mental health; depression detection algorithm multinomial naive bayes classification

Abstract

Community mental health is a serious challenge that must be encountered by a country. The results of the I-NAMHS (Indonesia-National Adolescent Mental Health Survey) show that one in three teenagers in Indonesia has mental health problems, while one in twenty teenagers in Indonesia has had a mental disorder in the last 12 months. Based on the results of a study from the Mayo Clinic in 2019, it was revealed that there is a phenomenon regarding the impact of social media on teenagers aged 13–16 years, where using social media more than three times a day can have a negative influence on mental health. This research aims to develop an approach to detecting depressive symptoms in social media users through messages delivered via Twitter. The crawling data result is labeled according to the Patient Health Questionnaire-9 (PHQ-9) questionnaire with three categories of depression levels (defined by the American Psychiatric Association), namely mild, moderate, and severe. The classification algorithm used is Multinomial Naive Bayes with Frequency-Relevance Frequency (TF-RF) as the weight determination and Min-Max Scaler as the normalization data. The evaluation results show that the resulting model has an accuracy value of 87.6%. Using this model, we detected from 76483 tweets that 2385 accounts were categorized as having mild depression, 5069 accounts were in moderate depression, and 3636 accounts were in severe depression. Therefore, in general, this approach can be applied as a depression early warning system for social media platform users.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Achmad Maududie
AU  - Putri Armaini
AU  - Priza Pandunata
PY  - 2024
DA  - 2024/06/29
TI  - Depression Early Warning System Based on Social Media Using Multinomial Naïve Bayes Algorithm
BT  - Proceedings of the 2nd International Conference on Neural Networks and Machine Learning 2023 (ICNNML 2023)
PB  - Atlantis Press
SP  - 62
EP  - 71
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-445-7_8
DO  - 10.2991/978-94-6463-445-7_8
ID  - Maududie2024
ER  -