Online review analytics using word alignment model on Twitter data

Abstract

Opinions are vital in almost all human activities and are key influencers of online user's behavior. In this regard, when we need to make a decision, we often search for the opinions of others. Opinions and its related concepts such as sentiments and emotions are the subjects of study of sentiment analysis and opinion mining. The inception of mining opinions from online reviews coincide with the social media on the Web, e.g., reviews, discussion forums, blogs, micro blogs, Twitter, and social networks. Due to the large volume of opinionated data generated and available in digital forms, it become essential to mine such data and find opinions or sentiments in order to obtain the overall opinion on the topic i.e. products, politics, sports, education reforms, and so on. In this direction, it is proposed to design an algorithm which extracts opinion targets and opinion words using word alignment model for online reviews extracted from Twitter. An opinion target is defined as the topic about which users express their opinions. An opinion words are defined as the words that are used to express users' opinions. The aim of the project is to determine the thoughts of blog or review writer with respect to some topic or the overall contextual polarity of online reviews using word alignment model. The key objective of this project is to design an algorithm that predict opinion words and opinion targets for analyzing the market status of a product by mining user reviews posted in social networking site namely the Twitter. For evaluation, Benchmark Customer Review Dataset and real tweets dataset pertaining to product reviews extracted using Twitter API. The results are measured in terms of Precision and Recall for accuracy of finding opinion words and targets in order to be an essential ingredient for opinion mining and sentiment analysis. The experimental results show that the proposed method achieves better accuracy in an efficient way. In addition, the ultimate outcome of this project to make choice of designing potential consumer oriented products e.g. Mobile, laptop, and so on.