Laisa Usrini, Intan (2025) Analisis Sentimen Terhadap Isu Kesehatan Mental Di Kalangan Remaja Menggunakan Algoritma Support Vector Machine (svm) Di Aplikasi X (twitter). Other thesis, Universitas Islam Riau.

Text 203510647.pdf - Submitted Version Restricted to Registered users only Download (1MB) | Request a copy

Abstract

Mental health among adolescents has become a growing global concern, as they face increasing social, academic and technological pressures. Social media, especially applications such as Twitter (X), are platforms where adolescents often share their feelings and experiences regarding mental health issues. This study aims to analyze the sentiment of X users towards mental health issues among teenagers using tweets uploaded on the X platform. In this research, tests were conducted by comparing Machine Learning algorithms namely Support Vector Machine (SVM) and Deep Learning algorithms namely Long Short Term Memory (LSTM) to classify positive, negative, and neutral sentiments. The dataset used amounted to 5470 with the number of negative sentiments 2166, neutral sentiments 1817, and positive sentiments 1487. Data was trained and tested using a ratio of 80:20, 70:30, 60:40 using both algorithms. The highest accuracy result is found in SVM with a ratio of 80:20 which is 92.4%. The LSTM was trained using the Word2vec approach using Skip-Gram and Continuous Bag of Words (CBOW) with dimensions of 100 and 200, the highest results were in Skip-Gram dimension 200 with an accuracy of 92.1%.

Actions (login required)

View Item