8 - Sentiment Analysis for Airlines Services Based on Twitter Dataset

Abstract

Opinion mining has been ordinarily connected with the examination of a content string to decide if a corpus is of a negative or positive sentiment. As of late, opinion mining has been stretched out to address issues, for example, recognizing objective from subjective suggestions and deciding the sources and points of various suppositions communicated in text informational collections, for example, tweets, message board, web blogs, movie reviews, and news. Companies can use sentiment extremity and opinion point acknowledgment to pick up a more profound comprehension and the general extent of estimations. These experiences can progress focused insight, enhance client benefit, accomplish better brand picture, and upgrade competitiveness. In the aircraft service industry, it is hard to gather information about clients’ input by polls, yet Twitter gives a sound information source to them to do client opinion examination. This paper presents positive, negative sentiment, and their correlation about customer tweets. BIRCH clustering and Association rule mining have been used in this chapter to get inside the dataset and retrieve hidden knowledge.