5.7　本章小结

本章讲解了关联规则、朴素贝叶斯、聚类这3类基础机器学习算法用于个性化推荐的理论知识，同时从算法原理和工程实现的角度简单总结了YouTube和Google News的三篇分别采用关联规则、朴素贝叶斯、聚类思路来做推荐的论文。这几篇论文有很强的工程指导意义，值得读者学习。

虽然这些算法原理简单、容易理解，但是这些算法却在工业界有过非常好的应用，在当时算是非常优秀的算法。这些算法现在可能看起来太简单了，也可能不会用在现在的推荐系统上，但它们朴素的思想下面蕴含的是深刻的道理，值得推荐从业者学习、思考、借鉴，希望读者可以很好地理解它们，并吸收这些朴素思想背后的精华。

参考文献

[1]James Davidson，Benjamin Liebald，Junning Liu，et al.The YouTube Video Recommendation System[C].[S.l.]:RecSys，2010.

[2]Jiahui Liu，Peter Dolan，Elin Rønby Pedersen.Personalized News Recommendation Based on Click Behavior[C].[S.l.]:Proceedings of the 2010 International Conference on Intelligent User Interfaces，2010.

[3]Abhinandan Das，Mayur Datar，Ashutosh Garg，et al.Google news personalization:Scalable online collaborative flitering[C].[S.l.]:WWW，2007.

[4]J J Sandvig，Bamshad Mobasher，Robin Burke.Robustness of collaborative recommendation based on association rule mining[C].[S.l.]:ACM，2007.

[5]Badrul Sarwar，George Karypis，Joseph Konstan，et al.Analysis of recommendation algorithms for e-commerce[C].[S.l.]:ACM，2000.

[6]Rakesh Agrawal，Ramakrishnan Srikant.Fast algorithms for mining association rules[C].[S.l.]:VLDB Conference，1994.

[7]Weiyang Lin，Sergio A Alvarez，Carolina Ruiz.Efficient adaptive-support association rule mining for recommender systems[C].[S.l.]:Data Mining and Knowledge Discovery，2002.

[8]Xiaobin Fu，Jay Budzik，Kristian J Hammond.Mining navigation history for recommendation[C].[S.l.]:Proceedings of the 5th international conference on Intelligent user interfaces，2000.

[9]Mei-Ling Shyu，C Haruechaiyasak，Shu-Ching Chen，et al.Collaborative filtering by mining association rules from user acess sequences[C].[S.l.]:Web Information Retrieval and Integration，2005.

[10]Jiawei Han，Jian Pei，Yiwen Yin，et al.Mining frequent patterns without candidate generation[C].[S.l.]:Data Mining and Knowledge Discovery，2004.

[11]Haoyuan Li，Yi Wang，Dong Zhang，et al.Pfp:parallel fp-growth for query recommendation[C].[S.l.]:RecSys，2008.

[12]Jian Pei，Jiawei Han，B Mortazavi-Asl，et al.Mining sequential patterns by pattern-growth:the PrefixSpan approach[C].[S.l.]:IEEE，2004.

[13]John S Breese，David Heckerman，Carl Kadie.Empirical analysis of predictive algorithms for collaborative filtering[C].[S.l.]:Proceedings of the Fourteenth conference on Uncertainty in artificial intelligence，1998.

[14]Yung-Hsin Chen，Edward I George.A Bayesian model for collaborative filtering[C].[S.l.]:AISTATS，1999.

[15]Koji Miyahara，Michael J Pazzani.Collaborative filtering with the simple Bayesian classifier[C].[S.l.]:PRICAI，2000.

[16]Kai Yu，Anton Schwaighofer，Volker Tresp，et al.Probabilistic memory-based collaborative filtering[C].[S.l.]:IEEE，2004.

[17]LH Ungar，DP Foster.Clustering methods for collaborative filtering[C].[S.l.]:AAAI workshop on recommendation systems，1998.

[18]Sonny Han Seng Chee，Jiawei Han，Ke Wang.Rectree:A efficient collaborative filtering method[C].[S.l.]:Data Warehousing and Knowledge Discovery，2001.

[19]Gui-Rong Xue，Chenxi Lin，Qiang Yang，et al.Scalable collaborative filtering using cluster-based smoothing[C].[S.l.]:SIGIR，2005.

[20]T Hofmann，J Puzicha.Latent class models for collaborative filtering[C].[S.l.]:IJCAI，1999.