A proposed system for modifying aspect based opinion mining for ranking of products

Abstract

The increasing use of the Internet has led to the dissemination of word-of-mouth through social networking sites. A product that has many favorable reviews are preferred in comparison with a similar kind of product with few favorable reviews. These feedback are rich in knowledge but are unstructured since they are written by non-experts. Aspect based Opinion Mining is modified in this proposed system to help consumers and enterprises in the analysis of user reviews by determining the strengths and weaknesses of the products. The aim of the project is to rank the product and its important aspects which would eventually help in faster decision-making. The developed system includes four main phases: Pre-processing Phase, Enhanced Aspect Identification and Opinion Word Extraction Phase, Aspect Polarity Identification Phase, Products and Aspects Ranking Phase. The developed system will be beneficial for the consumers and business analysts who explore user reviews for determining purchase decisions and enterprise strategies.