TV Program Recommendation for Multiple Viewers Based on user Profile Merging

  • Authors:
  • Zhiwen Yu;Xingshe Zhou;Yanbin Hao;Jianhua Gu

  • Affiliations:
  • School of Computer Science, Northwestern Polytechnical University, Xi Ancity, P.R. China;School of Computer Science, Northwestern Polytechnical University, Xi Ancity, P.R. China;Management School, Northwestern Polytechnical University, Xi Ancity, P.R. China;School of Computer Science, Northwestern Polytechnical University, Xi Ancity, P.R. China

  • Venue:
  • User Modeling and User-Adapted Interaction
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

Since today's television can receive more and more programs, and televisions are often viewed by groups of people, such as a family or a student dormitory, this paper proposes a TV program recommendation strategy for multiple viewers based on user profile merging. This paper first introduces three alternative strategies to achieve program recommendation for multiple television viewers, discusses, and analyzes their advantages and disadvantages respectively, and then chooses the strategy based on user profile merging as our solution. The selected strategy first merges all user profiles to construct a common user profile, and then uses a recommendation approach to generate a common program recommendation list for the group according to the merged user profile. This paper then describes in detail the user profile merging scheme, the key technology of the strategy, which is based on total distance minimization. The evaluation results proved that the merging result can appropriately reflect the preferences of the majority of members within the group, and the proposed recommendation strategy is effective for multiple viewers watching TV together.