Article Per Year (5 Year)
p-Index From 2021 - 2026
0.61
P-Index
This Author published in this journals
All Journal Jurnal Ilmiah Rekayasa dan Manajemen Sistem Informasi Building of Informatics, Technology and Science Malcom: Indonesian Journal of Machine Learning and Computer Science
Alfian, Zhevin
Unknown Affiliation
Computer Science & IT Engineering Other

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search
Journal : Building of Informatics, Technology and Science

 1 
Penerapan Support Vector Machine untuk Analisis Sentimen Pengguna X terhadap IndiHome, Biznet, dan Starlink Alfian, Zhevin; Afdal, M; Novita, Rice; Zarnelly, Zarnelly
Building of Informatics, Technology and Science (BITS) Vol 7 No 2 (2025): September 2025
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v7i2.7429

Abstract

This study aims to analyze user sentiment on the social media platform X toward three major internet service providers in Indonesia, IndiHome, Biznet, and Starlink. The analysis focuses on five key variables: internet speed, network stability, pricing and service packages, customer service quality, and coverage availability. A total of 4,500 data points were collected through data crawling, then processed using text mining techniques and the Support Vector Machine (SVM) algorithm, with data imbalance addressed through the Random Oversampling method. Evaluation results show that IndiHome consistently demonstrated the best performance, achieving an accuracy of up to 90% in the customer service quality variable, and an overall average accuracy above 85% across all variables. Biznet generally ranked second, with accuracy ranging from 63% to 80%. Starlink placed lowest overall, although it still recorded competitive results, such as 82% accuracy in the internet speed variable. The application of Random Oversampling improved the model’s classification accuracy by an average of 6–12% compared to the non-oversampling model. This study offers strategic insights into public perception of internet services and can serve as a reference for improving service quality based on data-driven user feedback.
Co-Authors Arif Marsal legito, Legito M. Afdal Rice Novita Sari, Windy Junita TEGUH ARIFIANTO, TEGUH Wulandari, Vina Zarnelly Zarnelly