Estimating Depressive Tendencies Of Twitter User Via Social Media Data

• Main Author: Loh, Hooi Teng
• Format: Undergraduates Project Papers
• Language: English
• Published: 2023
• Subjects: QA75 Electronic computers. Computer science
• Online Access: http://umpir.ump.edu.my/id/eprint/40189/1/CA19072.pdf; http://umpir.ump.edu.my/id/eprint/40189/

Nowadays, depression is a major mental health problem that affects people of all ages, genders, and ethnicities all over the world. People feel increasingly comfortable sharing their ideas on social media websites or applications practically every day in this age of contemporary communication and technology. The aim of this project is to study the properties of the text related to depressive tendencies via Twitter dataset. In this project will require huge dataset from twitter, so we will collect the Twitter dataset from Kaggle websites that already have the completed twitter dataset that can be downloaded in order to implement the estimating depressive tendencies of Twitter user. The twitter dataset can be used to test the level of depressive tendencies with three different machine learning algorithms. These three different machine learning algorithms which are Support Vector Machine, XGBoost, and Random Forest. We will use these three machine learning algorithms to compare the accuracy and performance of the depressed twitter user. Therefore, different machine learning have different types of features that can use to conduct the estimating depressive tendencies.