Modeling the Parameter Interactions in Ranking SVM with Low-Rank Approximation

doi:10.1109/TKDE.2018.2851257

	Modeling the Parameter Interactions in Ranking SVM with Low-Rank Approximation
	Xu, Jun 1; Zeng, Wei 2; Lan, Yanyan 2; Guo, Jiafeng 2; Cheng, Xueqi 2
	2019-06-01
发表期刊	IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
ISSN	1041-4347
卷号	31 期号:6 页码:1181-1193
摘要	Ranking SVM, which formalizes the problem of learning a ranking model as that of learning a binary SVM on preference pairs of documents, is a state-of-the-art ranking model in information retrieval. The dual form solution of a linear Ranking SVM model can be written as a linear combination of the preference pairs, i.e., w = Sigma((i,j)) alpha(ij) (x(i) - x(j)), where alpha(ij) denotes the Lagrange parameters associated with each preference pair (i, j). It is observed that there exist obvious interactions among the document pairs because two preference pairs could share a same document as their items, e.g., preference pairs (d(1), d(2)) and (d(1), d(3)) share the document d(1). Thus it is natural to ask if there also exist interactions over the model parameters alpha(ij), which may be leveraged to construct better ranking models. This paper aims to answer the question. We empirically found that there exists a low-rank structure over the rearranged Ranking SVM model parameters alpha(ij), which indicates that the interactions do exist. Based on the discovery, we made modifications on the original Ranking SVM model by explicitly applying low-rank constraints to the Lagrange parameters, achieving two novel algorithms called Factorized Ranking SVM and Regularized Ranking SVM, respectively. Specifically, in Factorized Ranking SVM each parameter alpha(ij) is decomposed as a product of two low-dimensional vectors, i.e., alpha(ij) = < v(i), v(j)>, where vectors v(i) and v(j) correspond to document i and j, respectively; In Regularized Ranking SVM, a nuclear norm is applied to the rearranged parameters matrix for controlling its rank. Experimental results on three LETOR datasets show that both of the proposed methods can outperform state-of-the-art learning to rank models including the conventional Ranking SVM.
关键词	Learning to rank ranking SVM parameter interactions low-rank approximation
DOI	10.1109/TKDE.2018.2851257
收录类别	SCI
语种	英语
资助项目	National Natural Science Foundation of China (NSFC)[61872338] ; National Natural Science Foundation of China (NSFC)[61773362] ; National Natural Science Foundation of China (NSFC)[61425016] ; National Natural Science Foundation of China (NSFC)[61472401] ; National Natural Science Foundation of China (NSFC)[61722211] ; Youth Innovation Promotion Association CAS[20144310] ; Youth Innovation Promotion Association CAS[2016102]
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Artificial Intelligence ; Computer Science, Information Systems ; Engineering, Electrical & Electronic
WOS记录号	WOS:000466933700012
出版者	IEEE COMPUTER SOC
引用统计	被引频次：9[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/4248
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Xu, Jun
作者单位	1.Renmin Univ China, Sch Informat, Beijing Key Lab Big Data Management & Anal Method, Beijing, Peoples R China 2.Chinese Acad Sci, Inst Comp Technol, CAS Key Lab Network Data Sci & Technol, Beijing, Peoples R China
推荐引用方式 GB/T 7714	Xu, Jun,Zeng, Wei,Lan, Yanyan,et al. Modeling the Parameter Interactions in Ranking SVM with Low-Rank Approximation[J]. IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING,2019,31(6):1181-1193.
APA	Xu, Jun,Zeng, Wei,Lan, Yanyan,Guo, Jiafeng,&Cheng, Xueqi.(2019).Modeling the Parameter Interactions in Ranking SVM with Low-Rank Approximation.IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING,31(6),1181-1193.
MLA	Xu, Jun,et al."Modeling the Parameter Interactions in Ranking SVM with Low-Rank Approximation".IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING 31.6(2019):1181-1193.