Recommending quality book reviews from heterogeneous websites

Purpose: In recent years, readers have limited amounts of time to pick the right books to read from a market that is filled with similar types of books. Aiming to read only good books, readers tend to check book reviews written by others. However, it is very difficult to find good book reviews. The aim of this paper is to present a book review recommendation system that collects reviews from heterogeneous sources on the Internet and performs quality judgments automatically. Users can then read the top-ranked reviews suggested by this recommendation system. Design/methodology/approach: In this paper, a book review recommendation system is constructed to collect, process, and judge the quality of book reviews from various heterogeneous sources. The quality measurement of book reviews uses review-evaluation techniques. The prediction results were validated with a ranking list produced by experts. Findings: The proposed system is effective and suitable for recommending quality book reviews from heterogeneous sources. The proposed quality measurement method is more effective than other more commonly used methods. Originality/value: This paper is one of the first to apply review evaluation techniques to the process of book review recommendation. The proposed system can collect and recognize book reviews from different websites with various forms of presentation. This evaluation shows that the quality measurement method produces better results than do other methods, such as ranking by rating score or by the date that the review was posted. Those methods are primarily used by commercial websites.