Aspect-Based Sentiment Classification Using Supervised Classifiers and Polarity Prediction Using Sentiment Analyzer for Mobile Phone Tweets

Twitter platform finds an important part in social marketing, election campaigns, academia and news. Sentiment analysis came into existence for finding polarities on different domains based on customer tweets online such as to check the opinions of people on a particular topic. Aspect-based sentiment analysis (ABSA) is a text analysis which identifies an entity or feature present in a given sentence and provides an opinion according to the given aspect. In this paper, we proposed an automatic approach for aspect sentiment classification and polarity prediction using mobile phone tweets. So, research contribution is given in the modules of tweets collection and preprocessing, implicit aspect term extraction towards sentiment classification and polarity prediction for mobile phone tweets. In order to predict sentiment polarity for aspect category, we use Vander sentiment analyzer and predict overall polarity as a compound value.