Support vector machine based emotional analysis of restaurant reviews
Leonard Gunawan, Maria Susan Anggreainy, Louis Wihan, Santy Santy, Gihon Yonathan Lesmana, Sabin Yusuf
Abstract
Most customers are frequently faced with the same question: "which restaurant serves the best food?" We employ sentiment analysis to address this question. Due to the high number of competitors in the restaurant industry, restaurant owners place a premium on customer feedback regarding the quality of the food and services they offer in order to increase customer satisfaction. Every word a customer utters carries a connotation and emotion. This procedure is essential for determining whether the customers' feelings toward our brand are positive, negative, or neutral. This study attempts to classify restaurant customer satisfaction in Jakarta using SVM (Support Vector Machine) and comparing it to the Nave Bayes classifier (NB). The results indicated a rise in SVM accuracy from 77% to 79%.