Selecting Best Features Using Combined Approach in POS Tagging for Sentiment Analysis?
Nikita D. Patel, Chetana Chand?
Journal Title:International Journal of Computer Science and Mobile Computing - IJCSMC
Today E-commerce popularity has made web an excellent source of gathering customer reviews / opinions about a product that they have purchased. The number of customer reviews that a product receives is growing at a very fast rate. Opinion mining from product reviews, forum posts and blogs is an important research topic today with many applications. Customers use the reviews for deciding quality of product to buy. So, opinion mining may be a Decision making process. It means reviews are given to promote or to demote the product. There is need to find how many reviews are positive and how many are negative. So, to find out it features for which classification is going to be performed should be best or optimal. This Paper presents various approaches of classification for sentiment analysis and proposed work is selecting best feature set such as pos tags from reviews which we can easily classify the review of customer. Only features which are giving best decision for analysis have been selected in pre-processing task and Combination of best feature set will be used to classify reviews.