Recovery of the Opinions Through the Specificities of Documents Text

Abstract

With the explosive growth of textual data from social media (social networks, forums, and blogs), the exploitation of these new sources of information has become crucial. Our work focuses on the sentiment analysis in this context of social media. To identify sentiments from messages, original text mining techniques must be proposed. This article presents a new method that integrates semantic and lexical knowledge for sentiment analysis. The proposed approach gives significant weight to the “sentiment words” for a classification task. Our study focuses on the 88milSMS corpus to highlight the specific characteristics of SMS data in the context of social media.