Analisis Sentimen Komentar Universitas di Indonesia Menggunakan Metode Naive Bayes dan SVM

Abstract

This research aims to analyze public sentiment towards Universitas Indonesia based on user reviews collected from Google Maps. In the era of digital information, online reviews serve as invaluable feedback channels, significantly influencing an institution's reputation and prospective student choices. This study employs a sentiment analysis approach to automatically classify reviews into positive, negative, and neutral categories. The methodology involves several key stages: data collection from Google Maps, comprehensive text preprocessing (including cleaning, tokenization, stopword removal, and stemming), and feature extraction using Term FrequencyInverse Document Frequency (TF-IDF). For classification, two prominent machine learning algorithms, Support Vector Machine (SVM) and Multinomial Naive Bayes, are utilized. Both models are trained and evaluated on the processed dataset to assess their performance in accurately classifying sentiment. A comparative analysis will be conducted to highlight the strengths and weaknesses of each algorithm in this specific context. The findings are expected to provide Universitas Indonesia with actionable insights into public perception, identify areas for improvement, and contribute to the understanding of sentiment analysis applications in educational contexts.