Abstract: With the rapid growth in ecommerce, reviews for popular products on the web have grown rapidly. Customers are often forced to wade through many online reviews in order to make an informed product choice. Scanning all of these reviews would be tedious, time consuming, boring and fruitless. It would be good if these reviews could be processed automatically and customers are provided with the limited generalized information. Opinion Mining plays a major role to summarize customer reviews and make it easy for online customers to determine whether to purchase the products or not.In this paper, we performed aspect level opinion mining on customer reviews using supervised learning algorithm. The proposed work performs the aspect-level opinion mining by extracting product aspects from reviews on ecommerce site and produces a summarized report of the most frequently discussed product aspects with regards to the number of appearances in positive, negative and neutral reviews. Here, frequent item set mining is used for aspect extraction and supervised learning algorithm (Support Vector Machine) is used to identify the number of the positive, negative and neutral opinion of each extracted aspect.
Keywords: Customer Reviews, Aspect Level Opinion Mining, Product Aspect Extraction, Supervised Learning Algorithm, Frequent Itemset Mining.