Perbandingan Metode Support Vector Machine dan Naïve Bayes dalam Analisis Sentimen Ulasan Pengguna terhadap Game Roblox

PAMUNGKAS, Emanuel Rahkito Haryo (2026) Perbandingan Metode Support Vector Machine dan Naïve Bayes dalam Analisis Sentimen Ulasan Pengguna terhadap Game Roblox. Undergraduate thesis, Universitas Katolik Widya Mandira.

	Text ABSTRAK.pdf Download (651kB)
	Text BAB I.pdf Download (349kB)
	Text BAB II.pdf Restricted to Repository staff only Download (388kB)
	Text BAB III.pdf Restricted to Repository staff only Download (217kB)
	Text BAB IV.pdf Restricted to Repository staff only Download (813kB)
	Text BAB V.pdf Download (340kB)
	Text DAFTAR PUSTAKA DAN SURAT BEBAS PLAGIAT.pdf Download (802kB)

Abstract

Roblox is a prominent online gaming platform in Indonesia that draws diverse user feedback. This study analyzes user sentiment regarding Roblox and evaluates the performance of Support Vector Machine (SVM) and Naïve Bayes (NB) algorithms in text-based sentiment analysis. The dataset consists of 20,000 Indonesian-language reviews collected via web scraping between August 12 and October 28, 2025, featuring a balanced distribution of short and long-form text. The data analysis process involves comprehensive text preprocessing—including Cleaning , case folding, tokenization, Stopword Removal, and normalization—followed by TF-IDF weighting and an 80:20 data split for training and testing. Model performance was measured using a confusion matrix, accuracy, precision, recall, and F1-score. The results demonstrate that SVM outperformed NB, achieving 81.11% accuracy and a 71.61% F1-score, compared to NB’s 70.39% accuracy and 44.81% F1-score. This research recommends the SVM algorithm for sentiment analysis of Indonesian game reviews and provides valuable insights for Roblox developers in understanding user perceptions.

Actions (login required)

View Item