Full-Text Search:
Home|Journal Papers|About CNKI|User Service|FAQ|Contact Us|中文
《Sciencepaper Online》 2011-04
Add to Favorite Get Latest Update

Learning to rank in heterogeneous network

Yang Zi,Tang Jie,Li Juanzi(Department of Computer Science and Technology,Tsinghua University,Beijing 100084,China)  
Massive heterogeneous resources have been currently available online,which brings richer information,along with more challenges.This paper concentrates on the issues of "learning to rank" in heterogeneous networks.First of all,the paper describes the investigation on topic-level random walk.We propose a three-step approach,especially focus on topic modeling of documents and the query and calculating a topic-level ranking score.Besides,we propose a general framework for heterogeneous cross-domain ranking,which simultaneously models the correlation between the source domain and the target domain,as well as learns the ranking models.We also develop an efficient EM-style solution,and discuss the generalized bound.The experiments show our proposed methods outperform the baseline methods.Finally,the thesis introduces the concept of specific expert search,which can be solved by the heterogeneous cross-domain ranking algorithm.
【Fund】: 高等学校博士学科点专项科研基金资助项目(20070003093);; 国家高技术研究发展计划(863计划)资助项目(2009AA01Z138)
【CateGory Index】: O223
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©2006 Tsinghua Tongfang Knowledge Network Technology Co., Ltd.(Beijing)(TTKN) All rights reserved