Scalable and interpretable product recommendations via overlapping co-clustering

被引:34
作者
Heckel, Reinhard [1 ]
Vlachos, Michail [2 ]
Parnell, Thomas [2 ]
Duenner, Celestine [2 ]
机构
[1] Univ Calif Berkeley, Berkeley, CA 94720 USA
[2] IBM Res Zurich, Zurich, Switzerland
来源
2017 IEEE 33RD INTERNATIONAL CONFERENCE ON DATA ENGINEERING (ICDE 2017) | 2017年
基金
欧洲研究理事会;
关键词
COMMUNITY STRUCTURE; EXPLANATIONS; FACTORIZATION;
D O I
10.1109/ICDE.2017.149
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
We consider the problem of generating interpretable recommendations by identifying overlapping co-clusters of clients and products, based only on positive or implicit feedback. Our approach is applicable on very large datasets because it exhibits almost linear complexity in the input examples and the number of co-clusters. We show, both on real industrial data and on publicly available datasets, that the recommendation accuracy of our algorithm is competitive to that of state-of-art matrix factorization techniques. In addition, our technique has the advantage of offering recommendations that are textually and visually interpretable. Finally, we examine how to implement our technique efficiently on Graphical Processing Units (GPUs).
引用
收藏
页码:1033 / 1044
页数:12
相关论文
共 51 条
[1]  
[Anonymous], [No title captured]
[2]  
[Anonymous], P ADV NEUR INF PROC
[3]  
[Anonymous], 1999, Athena scientific Belmont
[4]  
[Anonymous], 2016, P 25 ACM INT S HIGH
[5]  
[Anonymous], 2013, P 6 ACM INT C WEB SE, DOI [DOI 10.1145/2433396.2433471, 10.1145/2433396.2433471]
[6]  
[Anonymous], 2005, P 11 ACM SIGKDD INT
[7]  
[Anonymous], 2008, EUROPEAN C INFORM RE
[8]  
[Anonymous], 2011, CUDA by Example: An Introduction to General-Purpose GPU Programming
[9]  
[Anonymous], 2003, Proceedings of the ACM SIGKDD Conference on Knowledge Discovery and Data Mining
[10]  
[Anonymous], [No title captured]