Evaluating Recommender Behavior For New Users

被引:36
作者
Kluver, Daniel [1 ]
Konstan, Joseph A. [1 ]
机构
[1] Univ Minnesota, GroupLens Res, Dept Comp Sci & Engn, Minneapolis, MN 55455 USA
来源
PROCEEDINGS OF THE 8TH ACM CONFERENCE ON RECOMMENDER SYSTEMS (RECSYS'14) | 2014年
基金
美国国家科学基金会;
关键词
Recommender Systems; Evaluation; Profile Size; New User Experience; New User Problem; User Cold Start;
D O I
10.1145/2645710.2645742
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The new user experience is one of the important problems in recommender systems. Past work on recommending for new users has focused on the process of gathering information from the user. Our work focuses on how different algorithms behave for new users. We describe a methodology that we use to compare representatives of three common families of algorithms along eleven different metrics. We find that for the first few ratings a baseline algorithm performs better than three common collaborative filtering algorithms. Once we have a few ratings, we find that Funk's SVD algorithm has the best overall performance. We also find that ItemItem, a very commonly deployed algorithm, performs very poorly for new users. Our results can inform the design of interfaces and algorithms for new users.
引用
收藏
页码:121 / 128
页数:8
相关论文
共 17 条
  • [1] [Anonymous], 2008, INTRO INFORM RETRIEV, DOI DOI 10.1017/CBO9780511809071
  • [2] Bellogin Alejandro, 2012, THESIS
  • [3] Consolandi C, 2012, OMICS TECHNOLOGIES: TOOLS FOR FOOD SCIENCE, P13
  • [4] Cremonesi P., 2010, RECSYS 10
  • [5] Drenner S., 2008, RECSYS 08
  • [6] Ekstrand M., 2012, RECSYS 12
  • [7] Ekstrand M. D., 2011, RECSYS 11
  • [8] Funk S., 2006, NETFLIX UPDATE TRY T
  • [9] Golbandi N., 2011, WSDM 11
  • [10] Evaluating collaborative filtering recommender systems
    Herlocker, JL
    Konstan, JA
    Terveen, K
    Riedl, JT
    [J]. ACM TRANSACTIONS ON INFORMATION SYSTEMS, 2004, 22 (01) : 5 - 53