PROJECT TITLE :

Supervised and Unsupervised Aspect Category Detection for Sentiment Analysis with Co-Occurrence Data - 2017

ABSTRACT:

Using on-line consumer reviews as electronic word of mouth to help purchase-decision making has become increasingly in style. The Net provides an intensive source of shopper reviews, but one will hardly browse all reviews to obtain a truthful analysis of a product or service. A text processing framework which will summarize reviews, would therefore be desirable. A subtask to be performed by such a framework would be to search out the overall facet classes addressed in review sentences, for that this paper presents two strategies. In contrast to most existing approaches, the primary method presented is an unsupervised methodology that applies association rule mining on co-occurrence frequency knowledge obtained from a corpus to find these aspect categories. Whereas not on par with state-of-the-art supervised methods, the proposed unsupervised method performs better than several simple baselines, a similar but supervised technique, and a supervised baseline, with an F1-score of 67p.c. The second technique may be a supervised variant that outperforms existing ways with an F1-score of 84percent.