Combining the evidence of different relevance feedback methods for information retrieval |
| |
| Authors: | Joon Ho Lee |
| |
| Institution: | School of Computing, Soongsil University, 1-1 Sangdo-dong, Dongjak-gu, Seoul 156-743, South Korea |
| |
| Abstract: | It has been known that retrieval effectiveness can be significantly improved by combining multiple evidence from different query or document representations, or multiple retrieval techniques. In this paper, we combine multiple evidence from different relevance feedback methods, and investigate various aspects of the combination. We first generate multiple query vectors for a given information problem in a fully automatic way by expanding an initial query vector with various relevance feedback methods. We then perform retrieval runs for the multiple query vectors, and combine the retrieval results. Experimental results show that combining the evidence of different relevance feedback methods can lead to substantial improvements of retrieval effectiveness. |
| |
| Keywords: | information retrieval data fusion relevance feedback method |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
