ANALISIS SENTIMEN ULASAN MIE GACOAN SOLO VETERAN DI GOOGLE MAPS MENGGUNAKAN ALGORITMA NAIVE BAYES

Tariska Zidny Fatikhah (Universitas Duta Bangsa Surakarta)
Moh Muhtarom
Intan Oktaviani

Zonasi: Jurnal Sistem Informasi
Vol. 6 No. 3 (2024): Publikasi artikel ZONAsi: Jurnal Sistem Informasi Periode September 2024

Mie Gacoan Solo Veteran is one of the fastfood restaurant branches that is very popular with various groups, so that quite a few people have left an assessment of it. This research aims to help the Mie Gacoan Solo Veteran company in its efforts to understand consumer responses to their trademark and how people assess the products presented based on opinions and reviews on the Google Maps platform. Data was collected through a crawling process, then labeled and preprocessed before being extracted using the Term Frequency-Inverse Document Frequency (TF-IDF) technique and modeled with the Naive Bayes algorithm. The test results show that on unbalanced data, the model obtains an accuracy of 86%. However, after addressing data imbalance through an oversampling method using Synthetic Minority Over-sampling Technique (SMOTE), the model accuracy increased to 91%.