作者: Heungmo Ryang , Unil Yun
DOI: 10.1007/978-3-662-45402-2_25
关键词:
摘要: With the advent and growth of online bookstores, importance book reviews has been increased due to difficulty browse actual contents books in advance. This led necessity adequate systems for retrieving useful reviews. Recently, ranking algorithms have proposed this purpose. However, most them cannot consider discussions among users through replies. User discussion is an essential factor since it provides additional information. Although a state-of-theart algorithm with consideration proposed, deals only view reviewers’ participation. In paper, we propose estimating quality on basis all engaging discussions. Experimental results show that our outperforms state-of-the-art one terms precision, recall, NDCG.