Automatic combination of multiple ranked retrieval systems
SIGIR '94 Proceedings of the 17th annual international ACM SIGIR conference on Research and development in information retrieval
Combining the evidence of multiple query representations for information retrieval
TREC-2 Proceedings of the second conference on Text retrieval conference
Method combination for document filtering
SIGIR '96 Proceedings of the 19th annual international ACM SIGIR conference on Research and development in information retrieval
Predicting the performance of linearly combined IR systems
Proceedings of the 21st annual international ACM SIGIR conference on Research and development in information retrieval
Selecting the n-top retrieval result lists for an effective data fusion
CICLing'10 Proceedings of the 11th international conference on Computational Linguistics and Intelligent Text Processing
Hi-index | 0.00 |
Information retrieval researchers have long appreciated the value of combining, or fusing, multiple retrieval systems' relevance scores for a set of documents to improve retrieval performance. However, it is only recently that researchers have begun to consider adjusting the score fusion method to the user's topic and initial results. This study explores the value of fusing multiple retrieval systems' scores in a manner that adjusts to: the semantic and syntactic features of the user's natural language query, the various systems' biases toward long or short documents, and the extent to which the scores produced by the multiple systems are statistically independent.