推荐算法学习资料总结

 7emsp;本文综合整理了一些关于推荐算法的资料,资料来源注明在文章尾。

  • Books:

1.《推荐系统实践》项亮
  入门级教材,很薄,可以很快就看完,把很多基础而简单的问题讲的很详细。总体来说,此书性价比很高,值得入手一本研读
我买书喜欢上亚马逊, 因为亚马逊上很多都可以试读,这本书亚马逊就提供了试读,推荐大家先去试读下,再决定有没有购买价值。

2.《Recommender Systems Handbook》Paul B. Kantor
  有这本书就不用其它的了,很细很全,就是英文原版的有点小贵,真有志于做推荐系统的才去买吧,用到哪就翻书查。按人家的说法,所有敢自称handbook的书都是神书,没看过这本书出去吹牛逼时你都不好意思说自己是做推荐的。

3. Programming collective intelligence: building smart web 2.0 applications[M]
  寓教于乐的一本入门教材,附有可以直接动手实践的toy级别代码

4. Jannach D, Zanker M, Felfernig A, et al. Recommender systems: an introduction[M]. Cambridge University Press, 2010
  可以认为是2010年前推荐系统论文的综述集合

5. Celma O. Music recommendation and discovery[M]. Springer, 2010
  主要内容集中在音乐推荐,领域非常专注于音乐推荐,包括选取的特征,评测时如何考虑音乐因素。

6. Word sense disambiguation: Algorithms and applications[M]. Springer Science+ Business Media,2006
  如果涉及到关键词推荐,或是文本推荐, 则可以查阅该书。

  • Papers:

综述类:
  1. 2002 – Hybrid Recommender Systems: Survey and Experiments
  2. 2005 – Towards the Next Generation of Recommender Systems: A Survey of the State-of-the-Art and Possible Extensions。最经典的推荐算法综述
  3. 2009 – 个性化推荐系统的研究进展.周涛等
  4. 2011 – Collaborative Filtering Recommender Systems. JB Schafer 关于协同过滤最经典的综述
  5. 2012 – 项亮的博士论文《动态推荐系统关键技术研究》
  6. 2012 – Recommender systems L Lü, M Medo, CH Yeung, YC Zhang, ZK Zhang, T Zhou Physics Reports 519 (1), 1-49 (https://arxiv.org/abs/1202.1112
  7. 2017-基于深度学习的推荐系统_黄立威
协同过滤:

1.matrix factorization techniques for recommender systems. Y Koren
2.Using collaborative filtering to weave an information Tapestry. David Goldberg (协同过滤第一次被提出)
3.Item-Based Collaborative Filtering Recommendation Algorithms. Badrul Sarwar , George Karypis, Joseph Konstan .etl
4.Application of Dimensionality Reduction in Recommender System — A Case Study. Badrul M. Sarwar, George Karypis, Joseph A. Konstan etl
5.Probabilistic Memory-Based Collaborative Filtering. Kai Yu, Anton Schwaighofer, Volker Tresp, Xiaowei Xu,and Hans-Peter Kriegel
6.Recommendation systems:a probabilistic analysis. Ravi Kumar Prabhakar Raghavan.etl
7.Amazon.com recommendations: item-to-item collaborative filtering. Greg Linden, Brent Smith, and Jeremy York
8.Evaluation of Item-Based Top- N Recommendation Algorithms. George Karypis
9.Probabilistic Matrix Factorization. Ruslan Salakhutdinov
10.Tensor Decompositions,Alternating Least Squares and other Tales. Pierre Comon, Xavier Luciani, André De Almeida

基于内容的推荐:

1.Content-Based Recommendation Systems. Michael J. Pazzani and Daniel Billsus

基于标签的推荐:

1.Tag-Aware Recommender Systems: A State-of-the-Art Survey. Zi-Ke Zhang(张子柯), Tao Zhou(周 涛), and Yi-Cheng Zhang(张翼成)

推荐评估指标:

1、推荐系统评价指标综述. 朱郁筱,吕琳媛
2、Accurate is not always good:How Accuacy Metrics have hurt Recommender Systems
3、Evaluating Recommendation Systems. Guy Shani and Asela Gunawardana
4、Evaluating Collaborative Filtering Recommender Systems. JL Herlocker

推荐多样性和新颖性:
  1. Improving recommendation lists through topic diversification. Cai-Nicolas Ziegler
    Sean M. McNee, Joseph A.Konstan,Georg Lausen
  2. Fusion-based Recommender System for Improving Serendipity
  3. Maximizing Aggregate Recommendation Diversity:A Graph-Theoretic Approach
  4. The Oblivion Problem:Exploiting forgotten items to improve Recommendation diversity
  5. A Framework for Recommending Collections
  6. Improving Recommendation Diversity. Keith Bradley and Barry Smyth
推荐系统中的隐私性保护:

1.Collaborative Filtering with Privacy. John Canny
2.Do You Trust Your Recommendations? An Exploration Of Security and Privacy Issues in Recommender Systems. Shyong K “Tony” Lam, Dan Frankowski, and John Ried.
3.Privacy-Enhanced Personalization. Alfred Kobsa.etl
4.Differentially Private Recommender Systems:Building Privacy into the
Netflix Prize Contenders. Frank McSherry and Ilya Mironov Microsoft Research,
Silicon Valley Campus
5.When being Weak is Brave: Privacy Issues in Recommender Systems. Naren Ramakrishnan, Benjamin J. Keller,and Batul J. Mirza

推荐冷启动问题:

1.Tied Boltzmann Machines for Cold Start Recommendations. Asela Gunawardana.etl
2.Pairwise Preference Regression for Cold-start Recommendation. Seung-Taek Park, Wei Chu
3.Addressing Cold-Start Problem in Recommendation Systems. Xuan Nhat Lam.etl
4.Methods and Metrics for Cold-Start Recommendations. Andrew I. Schein, Alexandrin P opescul, Lyle H. U ngar

bandit(老虎机算法,可缓解冷启动问题):

1.Bandits and Recommender Systems. Jeremie Mary, Romaric Gaudel, Philippe Preux
2.Multi-Armed Bandit Algorithms and Empirical Evaluation
基于社交网络的推荐:

  1. Social Recommender Systems. Ido Guy and David Carmel
  2. A Social Networ k-Based Recommender System(SNRS). Jianming He and Wesley W. Chu
  3. Measurement and Analysis of Online Social Networks.
  4. Referral Web:combining social networks and collaborative filtering
    基于知识的推荐:
    1.Knowledge-based recommender systems. Robin Burke
    2.Case-Based Recommendation. Barry Smyth
    3.Constraint-based Recommender Systems: Technologies and Research Issues. A. Felfernig. R. Burke
    其他:
    Trust-aware Recommender Systems. Paolo Massa and Paolo Avesani
    推荐几篇对工业界比较有影响的论文吧:
  5. The Wisdom of The Few 豆瓣阿稳在介绍豆瓣猜的时候极力推荐过这篇论文,豆瓣猜也充分应用了这篇论文中提出的算法;
  6. Restricted Boltzmann Machines for Collaborative Filtering 目前Netflix使用的要推荐算法之一;
  7. Factorization Meets the Neighborhood: a Multifaceted Collaborative Filtering Model 这个无需强调重要性,LFM几乎应用到了每一个商业推荐系统中;
  8. Collaborative Filtering with Temporal Dynamics 加入时间因素的SVD++模型,曾在Netflix Prize中大放溢彩的算法模型;
  9. Context-Aware Recommender Systems 基于上下文的推荐模型,现在不论是工业界还是学术界都非常火的一个topic;
  10. Toward the next generation of recommender systems 对下一代推荐系统的一个综述;
  11. Item-Based Collaborative Filtering Recommendation Algorithms 基于物品的协同过滤,Amazon等电商网站的主力模型算法之一;
  12. Information Seeking-Convergence of Search, Recommendations and Advertising 搜索、推荐和广告的大融合也是未来推荐系统的发展趋势之一;
  13. Ad Click Prediction: a View from the Trenches 可以对推荐结果做CTR预测排序;
  14. Performance of Recommender Algorithm on top-n Recommendation Task TopN预测的一个综合评测,TopN现在是推荐系统的主流话题,可以全部实现这篇文章中提到的算法大概对TopN有个体会;
  15. http://dsec.pku.edu.cn/~jinlong/publication/wjlthesis.pdf 北大一博士对Netflix Prize算法的研究做的毕业论文,这篇论文本身对业界影响不大,但是Netflix Prize中运用到的算法极大地推动了推荐系统的发展;
    通过这些论文可以对推荐系统有个总体上的全面认识,并且能够了解一些推荐系统的发展趋势。剩下的就是多实践了。

推荐两篇必看(最好能自己实现)论文, 其他的论文其实都是在这基础上build起来的。
http://Amazon.com Recommendations Item-to-Item Collaborative Filtering
http://www.cin.ufpe.br/~idal/rs/Amazon-Recommendations.pdf

MATRIX FACTORIZATION TECHNIQUES FOR RECOMMENDER SYSTEMS
https://datajobs.com/data-science-repo/Recommender-Systems-%5BNetflix%5D.pdf
推荐两篇必看(最好能自己实现)论文, 其他的论文其实都是在这基础上build起来的。
http://Amazon.com Recommendations Item-to-Item Collaborative Filtering
http://www.cin.ufpe.br/~idal/rs/Amazon-Recommendations.pdf

MATRIX FACTORIZATION TECHNIQUES FOR RECOMMENDER SYSTEMS
https://datajobs.com/data-science-repo/Recommender-Systems-%5BNetflix%5D.pdf

  • Articles:

基于用户投票的排名算法(一):Delicious和Hacker News http://www.ruanyifeng.com/blog/2012/02/ranking_algorithm_hacker_news.html

youtube的推荐算法经历过好几次大的改动,都有论文发表的:https://www.zhihu.com/question/20829671

Netflix推荐算法,让每个人看到不一样的电影海报:https://juejin.im/post/5a2e71e351882575d42f5651

淘宝网的推荐算法:https://www.zhihu.com/question/29108284

SvdFeature
LibFM
Mahout
MLib

reference:

    原文作者:肥了个大西瓜
    原文地址: https://www.jianshu.com/p/759f806d0a59
    本文转自网络文章,转载此文章仅为分享知识,如有侵权,请联系博主进行删除。
点赞