论文部分内容阅读
为满足用户查询多样化需求,检索结果的多元化成为信息检索领域一个新的研究热点。文章提出一种支持搜索结果多元化的数据融合技术,采用权威的TREC文本检索会议(针对Web检索多元化任务的)数据,对提出的方法进行了实验验证,结果表明所提出的数据融合方法有效地提高了多元化性能,融合结果显著地优于最佳的成员检索系统,实验表明运用数据融合技术可以在由当今最先进的检索技术所产生的结果基础之上,进一步提高检索性能,该算法有较高的应用价值。
In order to meet the diverse needs of users, diversification of search results has become a new research hotspot in information retrieval. This paper presents a data fusion technology that supports the diversification of search results and validates the proposed method using the authoritative TREC text retrieval conference (for multi-tasking Web retrieval task) data. The results show that the proposed data fusion method is effective Improve the diversification performance, the fusion result is significantly better than the best member retrieval system, experiments show that the use of data fusion technology can be based on the results of today’s most advanced search technology to further improve the retrieval performance, the algorithm Have higher application value.