Abstract—The large amount of multimedia contents often cause the problem of information overload. To tackle this problem, it is necessary to develop personalization techniques to recommend most suitable contents to users. In this work, we develop a new social tag-based method for the recommendation of multimedia items, and compare it with several often-used methods. A context-aware platform is also implemented that takes into account different environment situations in order to make the most sensible recommendations. It has generally been agreed that the collaboration-based approach can provide better performance than the content-based approach. Yet, with the current trend of organizing and sharing digital content through user-created metadata (i.e., social tags), the performance and effectiveness of collaborative recommendation can be improved by using such metadata to further recognize how the users likes specific items. Social tags are brief descriptions of items and they are freely supplied by a community of internet users to aid the access of large collections of media (5)(6). The use of social tags provides an interesting alternative to current efforts at semantic web ontologies in content annotation (7) (8). As tagging is neither exclusive nor hierarchical and therefore can in some circumstances have an advantage over hierarchical taxonomies. In this work, we adopt this way for multimedia annotation, use the tag information to analyze how the user likes specific items, and exploit such user
[1]
Marios C. Angelides,et al.
Multimedia content modeling and personalization
,
2003,
IEEE MultiMedia.
[2]
Wei-Po Lee,et al.
A user-centered remote control system for personalized multimedia channel selection
,
2004,
IEEE Transactions on Consumer Electronics.
[3]
Giuseppe M. L. Sarnè,et al.
MUADDIB: A distributed recommender system supporting device adaptivity
,
2009,
TOIS.
[4]
Tong Zhang,et al.
Recommender Systems Using Linear Classifiers
,
2002
.
[5]
Yong Yu,et al.
Exploring social annotations for the semantic web
,
2006,
WWW '06.
[6]
George Macgregor,et al.
Collaborative tagging as a knowledge organisation and resource discovery tool
,
2006
.
[7]
Mark B. Sandler,et al.
Music Information Retrieval Using Social Tags and Audio
,
2009,
IEEE Transactions on Multimedia.
[8]
Bernardo A. Huberman,et al.
The Structure of Collaborative Tagging Systems
,
2005,
ArXiv.
[9]
Robin D. Burke,et al.
Hybrid Recommender Systems: Survey and Experiments
,
2002,
User Modeling and User-Adapted Interaction.
[10]
Wei-Po Lee,et al.
Deploying personalized mobile services in an agent-based environment
,
2007,
Expert Syst. Appl..
[11]
Matthias Baldauf,et al.
A survey on context-aware systems
,
2007,
Int. J. Ad Hoc Ubiquitous Comput..