Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Jeowandha Ria Wiyani
Fakultas Ilmu Komputer, Universitas Brawijaya
Computer Science & IT Control & Systems Engineering Education Electrical & Electronics Engineering Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Klasifikasi Stres berdasarkan Unggahan pada Media Sosial Twitter menggunakan Metode Support Vector Machine dan Seleksi Fitur Information Gain Jeowandha Ria Wiyani; Indriati Indriati; Sutrisno Sutrisno
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 6 No 12 (2022): Desember 2022
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Stress can happen to anyone and prolonged stress can cause mental health problems. However, many people continue to be unwilling to consult with mental health professionals about their concerns instead opting to complain on social media, such as Twitter. Many people use Twitter to vent their frustrations, making it possible to utilize text classification to determine someone's stress level from their tweets. In this work, the Support Vector Machine technique with Information Gain feature selection is used for text categorization. The data used in this study were 87 documents with details of 29 'Heavy' class documents, 29 'Medium' class documents, and 29 'Light' class documents. With a k value of 5, the test was run using the K-Fold Cross Validation method, and the distribution of training and test data was 80:20. The comparison of the results between the Support Vector Machine method alone with the combination of the Support Vector Machine and Information Gain methods produces the best accuracy on the Support Vector Machine method alone with an accuracy of 59.11%, precision of 29.99%, recall of 38.67%, and f-measure of 33.53%.
Co-Authors Indriati Indriati Sutrisno Sutrisno