Product Aspect Ranking Using Semantic Oriented Sentiment Classifier

Huge collections of consumer reviews are available on the Web expressing various opinions on multiple aspects of products.The important reviews are mostly not organized properly thereby creating problems in information navigation and knowledge acquisition. To address this problem, product aspect ranking is explored to automatically identify important product aspects or features from online consumer reviews. Probabilistic aspect ranking algorithm is proposed using K-NN based sentiment classifier. In this proposed system, semantic-oriented subjective information is first extracted and ranking is calculated based on the aspect frequency. This algorithm is experimented using T-Mobile dataset to show the effectiveness of the ranking approach. The scope of the proposed system is to organize the consumer review in appropriate way so that the product promotion can be done effectively based on reviews.