Recommendation over a Heterogeneous Social Network
Authors:
- Jing ZhangJZJing Zhang
- Jie TangJTJie Tang
- Bangyong LiangBLBangyong Liang
- Zi YangZYZi Yang
- Sijie WangSWSijie Wang
- Jingjing ZuoJZJingjing Zuo
- Juanzi LiJLJuanzi Li
- 1 July 2008
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 309-316
Abstract
With the Web content having been changed from homogeneity to heterogeneity, the recommendation becomes a more challenging issue. In this paper, we have investigated the recommendation problem on a general heterogeneous Web social network. We categorize the recommendation needs on it into two main scenarios: recommendation when a person is doing a search and recommendation when the person is browsing the information. We formalize the recommendation as a ranking problem over the heterogeneous network. Moreover, we propose using a random walk model to simultaneously ranking different types of objects and propose a pair-wise learning algorithm to learn the weight of each type of relationship in the model. Experimental results on two real-world data sets show that improvements can be obtained by comparing with the baseline methods.Keywords
This publication has 8 references indexed in Scilit:
- Social Network Extraction of Academic Researchers2007
- Expertise networks in online communities2007
- Object-level ranking2005
- Item-based top-Nrecommendation algorithmsACM Transactions on Information Systems, 2004
- Latent semantic models for collaborative filteringACM Transactions on Information Systems, 2004
- Studying Recommendation Algorithms by Graph AnalysisJournal of Intelligent Information Systems, 2003
- FabCommunications of the ACM, 1997
- Using collaborative filtering to weave an information tapestryCommunications of the ACM, 1992