Opinion mining: Book-ranking application

DOI

説明

A lot of products are sold on electronic commerce websites. There are many customer reviews available on the Internet. Users read them to select a product to purchase. However, going through the reviews is time-consuming because the number of reviews is large for each product. It is important to reflect reviews in a ranking to make a decision easily. The aim of this research is ranking products based on opinions of customers by proposing a new ranking method. We focused on the customer reviews on books. In this research, we collected all reviews on each book and classified them as positive or negative. After that, we calculated the rates of positive sentences and negative ones and made a ranking of books based on them. As a result from the experiment, we got a book ranking list. This ranking strongly depends on the number of customer reviews and their contents. This approach was illustrated reviews on books on Ruby on rails. In the future, we are planning to apply a machine learning algorithm to classify reviews more accurately.

収録刊行物

詳細情報 詳細情報について

問題の指摘

ページトップへ