Microsoft Research Asia at the NTCIR-10 Intent Task

Microsoft Research Asia at the NTCIR-10 Intent Task
复制标题

DOI:
--
复制
发表时间:
2013
期刊:
IEICE Trans. Inf. Syst.
影响因子:
--
通讯作者:
Jialong Han;Qinglei Wang;Naoki Orii;Zhicheng Dou;T. Sakai;Ruihua Song
Jialong Han;Qinglei Wang;Naoki Orii;Zhicheng Dou;T. Sakai;Ruihua Song
中科院分区:
其他
文献类型:
--
作者:
Jialong Han;Qinglei Wang;Naoki Orii;Zhicheng Dou;T. Sakai;Ruihua Song

文献摘要

相似文献

微软亚洲研究院参与了NTCIR - 10 INTENT任务的子主题挖掘子任务和文档排序子任务。在子主题挖掘子任务中,我们从查询建议、点击数据和查询的顶部结果中挖掘子主题,并根据它们对于给定查询的重要性对其进行排序。在文档排序子任务中,我们通过估计挖掘出的子主题的意图类型并结合多个搜索引擎的结果,使顶部搜索结果多样化。实验结果表明,就D(cid:2)-nDCG@10而言,我们在日语子主题挖掘方面表现最佳的一次运行在所有14次运行中排名第2。我们所有的日语文档排序运行都优于没有多样化的基线排序。
Microsoft Research Asia participated in the Subtopic Mining sub-task and Document Ranking subtask of the NTCIR-10 INTENT Task. In the Subtopic Mining subtask, we mine subtopics from query suggestions, clickthrough data and top results of the queries, and rank them based on their importance for the given query. In the Document Ranking subtask, we diversify top search results by estimating the intent types of the mined subtopics and combining multiple search engine results. Experimental results show that our best Japanese subtopic mining run is ranked No. 2 of all 14 runs in terms of D (cid:2) -nDCG@10. All of our Japanese document ranking runs outperform the baseline ranking without diversi fi cation.