Comparing Machine Learning Techniques for Sentiment Analysis

Abstract: Sentiment Analysis uses Natural Language Processing (NLP) and text analysis to systematically identify and extract subjective opinion of a document. There are several ways to evaluate the polarity of document. This paper gives insights on various machine classifiers used. Each Classifiers are evaluated separately using predefined metrics to find the best classifier for correctly determining the polarity of document.

Keywords: Sentiment Analysis, Polarity, Machine Learning Classifiers, confusion matrix, K-fold cross validation