Sentiment Analysis of Tiktok Shop Using Multinomial Naïve Bayes Method and BM25

Authors

  • Andrew Arjunanda Yasin Universitas Pembangunan Nasional "Veteran" Jawa Timur
  • Dwi Arman Prasetya
  • Tresna Maulana Fahrudin

Keywords:

TikTok Shop, BM25, Analisis Sentimen, Multinomial Naïve Bayes

Abstract

In the current digital era, TikTok has emerged as a popular application among internet users. One prominent feature is the TikTok Shop, which allows users to shop directly through the platform. However, on October 4, 2023, TikTok Shop was temporarily suspended by the government due to online trade regulation policies. After reopening on December 12, 2023, various responses from the public emerged in reaction to this phenomenon. This research conducted sentiment analysis on TikTok Shop by observing public responses on Twitter using the Multinomial Naïve Bayes classification method with word weighting using the BM25 and TF-IDF methods. The analysis results showed a majority of positive sentiment regarding TikTok Shop, indicating disapproval of the previous closure policy. In testing with a 10% test sample size, BM25 showed slightly better performance than TF-IDF.

Downloads

Download data is not yet available.

Additional Files

Published

28-08-2024

How to Cite

Yasin, A. A., Prasetya, D. A., & Fahrudin, T. M. (2024). Sentiment Analysis of Tiktok Shop Using Multinomial Naïve Bayes Method and BM25. Jurnal Ilmiah Teknologi Informasi Asia, 18(2), 24–31. Retrieved from https://jurnal.asia.ac.id/index.php/jitika/article/view/1004

Issue

Vol. 18 No. 2 (2024): Volume 18 nomor 2 2024 (8)

Section

Article

License

Copyright (c) 2024 Andrew Arjunanda Yasin, Dwi Arman Prasetya, Tresna Maulana Fahrudin

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

Upon acceptance for publication, authors transfer copyright of their article to Jurnal Ilmiah Teknologi Informasi Asia. This includes the rights to reproduce, transmit, and translate the material in any form or medium.

While the editorial board endeavors to ensure accuracy, they accept no responsibility for the content of articles or advertisements. Liability rests solely with the respective authors and advertisers.

Website material is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0). Under this license, users are free to share and adapt the material for any purpose, including commercial use, provided license terms are met. These freedoms are irrevocable by the licensor under such conditions.