Title: | a unified iterative optimization algorithm for query model and ranking refinement |
Author: | Huang Yunping
; Sun Le
; Nie Jian-Yun
|
Source: | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
|
Conference Name: | 6th Asia Information Retrieval Societies Conference, AIRS 2010
|
Conference Date: | 40878
|
Issued Date: | 2010
|
Conference Place: | Taipei, Taiwan
|
Keyword: | Algorithms
; Computational linguistics
; Information retrieval
; Infrared devices
; Iterative methods
; Optimization
; Search engines
; Vector spaces
|
Publish Place: | Germany
|
Indexed Type: | EI
|
ISSN: | 3029743
|
ISBN: | 3642171869
|
Department: | (1) Institute of Software, Chinese Academy of Sciences, Beijing, China; (2) Université de Montréal, Montréal, QC, Canada
|
Sponsorship: | National Science Council; Ministry of Education
|
English Abstract: | 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. |
Content Type: | 会议论文
|
URI: | http://ir.iscas.ac.cn/handle/311060/8636
|
Appears in Collections: | 基础软件国家工程研究中心_会议论文
|
File Name/ File Size |
Content Type |
Version |
Access |
License |
|
a unified iterative optimization algorithm for query model and ranking refinement.pdf(221KB) | -- | -- | 限制开放 | -- | 联系获取全文 |
|
Recommended Citation: |
Huang Yunping,Sun Le,Nie Jian-Yun. a unified iterative optimization algorithm for query model and ranking refinement[C]. 见:6th Asia Information Retrieval Societies Conference, AIRS 2010. Taipei, Taiwan. 40878.
|
|
|