Nomenclature of Diverse Feature Selection in Sentiment Analysis using Machine Learning Techniques: A Comparative Study

Sentiment analysis has gained in importance in recent decades as the amount of digital text documents has increased in tandem with the advent of IT. The goal of sentiment analysis in the field of natural language processing (NLP) is to extract positive or negative polarity from social media text. Feature Selection (FS) is important based on critical analysis and efficiently identifying the necessary information for better segregation. The purpose of this research is to assess and compare the performance of several sentiment analysis FS techniques. Several Feature Selection (FS) procedures are studied in order to select the optimal Selection of features from a feature set. Various Machine Learning techniques such as Decision Tree (DT), Naive Bayes (NB), Support Vector Machines (SVM), K-Nearest Neighbour (KNN) and Logistic Regression (LR) are used to train the selected features on the Online Movie Reviews Dataset. According to the findings of the experiments, the proposed system decreases data dimensionality, picks efficient aspects for better analysis