推荐系统论文

几篇对工业界比较有影响的论文

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;

Restricted Boltzmann Machines for Collaborative Filtering 目前Netflix使用的主要推荐算法之一;

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;

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;

Context-Aware Recommender Systems 基于上下文(情景)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;

Toward the next generation of recommender systems对下一代推荐系统的一个综述;

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;

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;

Ad Click Prediction: a View from the Trenches 可以对推荐结果做CTR预测排序;

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;

Netflix Prize 中的协同过滤算法 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;

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 

- Recommendations Item-to-Item Collaborative Filtering 

- MATRIX FACTORIZATION TECHNIQUES FOR RECOMMENDER SYSTEMS

当前推荐系统所面临的挑战相关研究

Current Challenges and Visions in Music Recommender Systems Research,Recsys2017录用的一篇文章,阐述了当前的音乐推荐系统所面临的挑战和远景。

AAAI 2018 录用推荐系统相关的部分Papers

ATRank: An Attention-Based User Behavior Modeling Framework for Recommendation

Attention-based Transactional Context Embedding for Next-Item Recommendation

Collaborative Filtering with Social Exposure: A Modular Approach to Social Recommendation

Compatibility Family Learning for Item Recommendation and Generation

Coupled Poisson Factorization Integrated with User/Item Metadata for Modeling Popular and Sparse Ratings in Scalable Recommendation

Exploiting Emotion on Reviews for Recommender Systems

你可能感兴趣的:(推荐系统论文)