USE Natural Language Processing TECHNIQUE FOR FEATURE SELECTION IN SENTIMENT ANALYSIS ON VIETNAMESE

Abstract

Sentiment analysis is a special form of text classification, in which a document is classified to predict the sentiment automatically as positive, negative or neutral. The process of sentiment analysis needs to deal with three basic problems: feature selection, identifying sentiment words and sentiment classification.

This article deals with the problem of feature selection, one of the important steps in the process of sentiment analysis. We analyze clearly the aspects in the approach of natural language processing, present the basic characteristics of Vietnamese, and then propose some approaches on Vietnamese to improve the efficiency of this feature selection process.