Абстрактный

Design of Review Based Product Aspect Ranking With Transaction Id Verification

Rupesh, K.sivaraman

A wide range of consumer reviews are available on different e-commerce websites for different products. Consumer reviews contain a rich set of knowledge which can be used for finding both product aspects or features as well as sentiment of consumer on different aspects of the product. The important features of any product can be extracted from the consumer review by finding the frequent noun terms in the reviews which are commented by large number of consumer, since the aspects are usually noun or noun phrases (Liu, 2009). Then we find the aspect - level sentiment word from the consumer review by using sentiment lexicon which consist a list of sentiment words. Then we develop a review based product aspect ranking framework which will use aspect frequency and the influence of consumer openions given to each and every product aspect for giving ranking to the product. Since we are giving ranking to the product based on the consumer review so we provide a unique transaction-id to every consumer on their successful transaction, Then we use Transaction-Id verification system to verify the customer and to allow them to write review.