高效个性化PageRank算法综述  

Efficient personalized PageRank computation: a survey

作　　者：朱凡微[1] 吴明晖[2] 应晶[1,2] 

机构地区：[1]浙江大学计算机学院,杭州310027 [2]浙江大学城市学院,杭州310015

出　　处：《中国科技论文》2012年第1期7-13,共7页China Sciencepaper

基　　金：清华-腾讯互联网创新技术联合实验室资助项目(2011-8)

摘　　要：对高效的个性化PageRank算法进行综述,从个性化程度、可扩展性、计算效率和精确度等方面对部分个性化算法、可扩展的PPV算法和混合算法等3类算法进行了详细分析和比较,并讨论了目前工作中的不足及未来的研究方向。As personalized PageRank has been widely leveraged for ranking on graph-structured scenarios, its computation efficiency becomes a prominent issue. We survey on an array of work that concentrate on efficient and scalable personalized PageRank computation, ranging from earlier work that attempt to use partial precomputation to improve online efficiency, to recent work that estimate approximate PPV for full peronalization and the hybrid methods. We compare these methods in terms of the ability of personalization, scalability, online/oitline efficiency and accuracy. We also point out a few possible research directions at the end of this paper.

关 键 词：WEB搜索 个性化排序 PAGERANK算法 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]