Supervised Machine Learning Approaches for Customer Reviews Sentiment Analysis

Sentiment analysis, also known as opinion mining, is a natural language processing (NLP) method to extract subjectivity and polarity from a body of text, i.e., identify whether the data is positive, negative, or neutral. In this paper, researchers have presented vectorization techniques like Bag of Words, TF-IDF, and HashingVectorizer along with the following text classification algorithms: K-Nearest Neighbour, Multinomial Naïve Bayes, Random Forest Classifier, Decision Tree and SVM. A comparative analysis is performed on the performance of different algorithms and vectorization technique combinations by computing statistical parameter like accuracy, specificity, sensitivity, false positive rate, false negative rate, negative predictive rate, false discovery rate, precision, recall and F1-score.