Abstract | ||
---|---|---|
The personalization of TV viewing has evolved, along with the proliferation of TV program content provided through various means, to become a key issue in future TV systems. It has thus become important to construct personalized user profiles for use in recommending TV programs in order to effectively personalize TV viewing. We have developed a method for automatically constructing a user profile on the basis of the user's viewing history. It creates a hierarchical tree structured graph into which a categorical data set from Wikipedia is transformed using a network clustering algorithm. We also developed a human-device interaction system and carried out evaluation experiments in which ten participants were watched TV programs in their everyday home environment. The results show that the proposed method is more effective than a convebtinal method. |
Year | DOI | Venue |
---|---|---|
2016 | 10.1109/BMSB.2016.7521929 | 2016 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB) |
Keywords | Field | DocType |
user profiles,network clustering,personalized TV,Wikipedia,user interest | Data structure,User profile,Categorical variable,Computer science,Interactive television,Encyclopedia,Multimedia,Electronic publishing,Personalization,The Internet | Conference |
ISSN | ISBN | Citations |
2155-5044 | 978-1-4673-9045-3 | 0 |
PageRank | References | Authors |
0.34 | 8 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Masahide Naemura | 1 | 97 | 12.91 |
Masaki Takahashi | 2 | 38 | 7.23 |
Simon Clippingdale | 3 | 61 | 8.98 |
Yuko Yamanouchi | 4 | 17 | 4.20 |
Hiroshi Fujisawa | 5 | 6 | 4.52 |