| a unified iterative optimization algorithm for query model and ranking refinement | |
| Huang Yunping; Sun Le; Nie Jian-Yun | |
| 2010 | |
| 会议名称 | 6th Asia Information Retrieval Societies Conference, AIRS 2010 |
| 会议录名称 | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
| 页码 | 514-526 |
| 会议日期 | 40878 |
| 会议地点 | Taipei, Taiwan |
| 收录类别 | EI |
| 出版地 | Germany |
| ISSN | 3029743 |
| ISBN | 3642171869 |
| 部门归属 | (1) Institute of Software, Chinese Academy of Sciences, Beijing, China; (2) Université de Montréal, Montréal, QC, Canada |
| 摘要 | Document ranking and query model estimation can be considered as optimization problems. In this paper, we propose an iterative algorithm for optimizing query model and ranking function simultaneously in the context of language model and vector space model, respectively. This algorithm extends the risk minimization framework by incorporating manifold structure of word graph and document graph, and it provides a unified formulation of several existing heuristics for document ranking and query modeling. Moreover, we extend our algorithm by incorporating users true feedback information, and derive a new ranking model. Experimental results on four TREC collections show that our model is effective. © 2010 Springer-Verlag. |
| 关键词 | Algorithms Computational Linguistics Information Retrieval Infrared Devices Iterative Methods Optimization Search Engines Vector Spaces |
| 主办者 | National Science Council; Ministry of Education |
| 内容类型 | 会议论文 |
| URI标识 | http://ir.iscas.ac.cn/handle/311060/8636 |
| 专题 | 基础软件国家工程研究中心 |
| 推荐引用方式 GB/T 7714 | Huang Yunping,Sun Le,Nie Jian-Yun. a unified iterative optimization algorithm for query model and ranking refinement[C]. Germany,2010:514-526. |
| 条目包含的文件 | ||||||
| 文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
| a unified iterative （221KB） | 开放获取 | -- | 请求全文 | |||
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。
修改评论