A Comparative Study of SentiWordNet and Other Lexicons for Social Media Sentiment Analysis

S. P.  Ponnusamy

Abstract

Another name for opinion mining is "Sentiment Analysis" (SA). Sentiment Analysis is a text classification algorithm that consistently advances the topic of study. Sentiment Analysis, or categorizing opinions into positive or negative categories, is indeed the primary goal of information extraction. Both information producers and consumers face significant challenges because of the explosion of information available online. For a business to be successful, it is crucial to comprehend the thoughts, opinions, and satisfaction of its customers. Sentiment analysis involves the electronic evaluation of individuals' feelings, emotions, and viewpoints. Utilizing social networking services, individuals can voice their views regarding various goods, occasions, or services. These reviewers overuse acronyms and jargon to convey their own opinions. As a result, word analysis is crucial for sentiment analysis. Thus, a database of words is needed for the purpose of sentiment analysis, and one such database is SentiWordNet derived from Wordnet dictionary. Sentiment analysis and sentiment are supported by the linguistic resource SentiWordNet. SentiWordNet is an opinion lexicon that was inferred as from WordNet directory, and every phrase is linked to something like a series of data total score that indicates whether a word is correlated to favourable or bad emotion. This study reviews some of the work carried out using the SentiWordNet database.

Abstract

Keywords