Content-boosted maximum margin matrix factorization for Flickr group recommendation

Authors:
Yilun Wang;Liang Chen;Jian Wu
Affiliations:
Zhejiang University, Hangzhou, China;Zhejiang University, Hangzhou, China;Zhejiang University, Hangzhou, China
Venue:
Proceedings of the 2013 workshop on Data-driven user behavioral modelling and mining from social media
Year:
2013

Citing 6
Cited 0

GroupLens: an open architecture for collaborative filtering of netnews

CSCW '94 Proceedings of the 1994 ACM conference on Computer supported cooperative work
Fast maximum margin matrix factorization for collaborative prediction

ICML '05 Proceedings of the 22nd international conference on Machine learning
Analyzing Flickr groups

CIVR '08 Proceedings of the 2008 international conference on Content-based image and video retrieval
Introduction to Information Retrieval

Introduction to Information Retrieval
Matrix Factorization Techniques for Recommender Systems

Computer
Modeling Flickr Communities Through Probabilistic Topic-Based Analysis

IEEE Transactions on Multimedia

Quantified Score

Hi-index	0.00

Visualization

Abstract

Recommending groups or communities to users can greatly improve the browsing experience in online photo sharing sites, e.g. Flickr. However, directly applying collaborative filtering techniques to group recommendation will suffer from "cold start" problem since many users do not affiliate to any groups. In this paper, we propose a hybrid recommendation method named Content-boosted Maximum Margin Matrix Factorization (CM3F), which combines collaborative user-group information with user similarity obtained from their uploaded images. Therefore, CM3F not only inherits the advantages of the state-of-the-art Maximum Margin Matrix Factorization (MMMF) method, but also owns the merits of the content-based graph regularization. The experiments conducted on our crawled dataset with 2196 users, 985 groups and 334467 images from Flickr demonstrate the effectiveness of our framework.