Restaurant Reviews Sentimental Analysis Using Machine Learning Approach

Two prominent machine learning methods applied to sentiment analysis of restaurant reviews they are Logistic Regression and Support Vector Machine (SVM). The dataset comprises 1001 restaurant reviews sourced from Kaggle, classified into positive and negative sentiments based on a numerical scale. Initially, Logistic Regression is used for binary classification and predicting the likelihood of a review being positive or negative. SVM is most widely used algorithms for supervised learning, both linear and non-linear variants, is subsequently employed to optimize classification accuracy by finding hyperplanes that separate classes with maximum margin in feature space. This study aims to compare the effectiveness of these algorithms in discerning sentiment from textual reviews, highlighting their respective strengths and performance outcomes. Experiment results show that the overall accuracy achieved for sentiment classification task using Logistic Regression model is 76.40% and 76.80% in Support Vector Machine.