FEATURE SELECTION IN TWITTER SENTIMENT ANALYSIS USING ENHANCED PESOFSA

Abstract

Sentiment analysis commencing Twitter subsist most concerning all exciting explore field in recent times. Sentiment analyses combine NLP (natural language processing) technique by all data mining approach used for structure similar system. The Present research, these introduce one competent classification designed for tweet sentiment mining. Enhanced PESOFSA methods build one machine learning method as detect positives also negatives tweet. That proposes in possible explanation whichever improve all stage about accurateness by excellent point in time efficient. In particular, this develops a different aspect grouping methods that utilize every emotion lexicon also extracted tweets unigrams connected with great information gain. A propose method be able to exist use being measure users’ sentiment taken away our tweet that act especially functional during various application such as thing advertising, political polarization recognition with product review.