Microsoft Research Asia at the NTCIR-9 Intent Task.
Jialong Han, Qinglei Wang, Naoki Orii, Zhicheng Dou, Tetsuya Sakai, Ruihua Song · NTCIR · 2011
In NTCIR-9, we participate in the Intent task, including both the Subtopic Mining subtask and the Document Rank- ing subtask. In the Subtopic Mining subtask, we mine sub- topics from query logs and top results of the queries, and rank them based on their relevance to the query and the similarity between them. In the Document ranking Subtask, we diversify top search results using the mined subtopics based on a general multi-dimensional diversification frame- work. Experimental results show that our best Chinese subtopic mining run is ranked No. 2 of all 42 runs in terms of DnDCG@10. Our Chinese document ranking runs gen- erally outperform other runs in terms of I-rec. Our best Chinese document ranking runs is the No. 4 of all 24 runs in terms of DnDCG@10. Our Japanese document rank- ing runs perform the best both in terms of D-nDCG and in terms of D-nDCG.