A Relevance Feedback Retrieval system based on indri toolkit

摘要：

Relevance feedback is an important application in information retrieval on Internet This paper introduces a relevance feedback retrieval system to improve the searching results. The system is built on the Indri toolkit, using pseudo relevance feedback method. First, we introduce the framework of the relevance feedback system and the methods we used in each module. The main module of the system is relevance feedback module. In this module, our algorithm called KNN-KL-LM algorithm is introduced in details for query expansion, which is essential for relevance feedback. Experiments show that the retrieval results are obviously improved.

关键词： Indri KNN Language Model KL-divengence

作者: Chun-Bo Liu Yan Li Wei-Ran Xu Si Li Jun Guo

作者单位: School of Information and Communication Beijing University of Posts and Telecommunications Beijing, China

会议类型: 国际会议

会议名称: 2011 2nd International Conference on Data Storage and Data Engineering(DSDE 2011)(2011年第二届数据存储与数据工程国际会议)

会议地点: 西安

会议语种:英文

页码: 4-7

在线出版日期: 2011-05-13（万方平台首次上网日期，不代表论文的发表时间）