Paper Title
A Comparative Analysis of Different Machine Learning Classification Models for Sentiment Analysis
Abstract
With the world's increasing reliance on digital technology, the production of textual documents has skyrocketed,
creating a need for proper organization and categorization. Text classification, also known as text categorization, is the
process of sorting text into distinct groups. In this study, we analyze the IMDB dataset of 50,000 movie reviews and design a
classification system. We compare Linear SVC, Bernoulli Naive Bayes, Logistic Regression, Multinomial Naive Bayes, and
Random Forest as classification algorithms for sentiment analysis and determining the polarity of the reviews. The classifiers
were analyzed and compared based on various parameters such as precision, accuracy, F1-score, recall, and confusion
matrix. The best machine learning algorithm for text sentiment analysis of the IMDB review dataset is determined by the
classifier that performs the best on these parameters.
Keywords - Text Classification∙ Sentiment Analysis∙ Machine Learning ∙ Logistic Regression∙ Random Forest∙
MultinomialNaïveBayes∙Bernoulli’s NaïveBayes∙LinearSupportVectorClassifier∙ Naturallanguageprocessing