Ranking book reviews based on user influence

Abstract

As the number of people who buy books online increases, it is becoming a way of life for people. Recently, algorithms to retrieve book reviews have been proposed for searching meaningful information since ranking algorithms for general purpose are not suitable. Although the previous algorithms consider features of book review, they calculate ranking scores without reflecting user influence. In this paper, thus, we propose a novel algorithm for ranking book reviews based on user influence. To apply user influence, the proposed algorithm uses recommendations by other users. For performance evaluation, we perform precision and recall tests. The experimental results show that the proposed algorithm outperforms previous algorithms for searching book reviews. © 2013 Springer Science+Business Media Dordrecht(Outside the USA).