Ranking of Products through Blog Analysis

Abstract

Many web blogs contain comments of users on different products. Such comments are often helpful for a naive user to decide which particular product (from among various alternatives available in the market) he/she is interested in. However, manual analysis of these blogs is timeconsuming. In this work we propose a tool for automatic analysis of blogs on different products, and rank them on the basis of certain key features. The task however is not straightforward as users’ comments are often replete with ungrammatical or poorly-structured sentences, incoherence of themes, usages of synonyms, and many other usual NLP problems, which need to be dealt with appropriately. The overall procedure takes the following steps: Preprocessing, POS tagging, Brand identification, Feature identification and Ranking. The scheme has been tested on blogs of Laptops and Automobiles. The initial results are found to be very promising.