Cross community news event summary generation based on collaborative ranking

  • Authors:

  • Chunxi Liu;Weigang Zhang;Shuqiang Jiang;Qingming Huang

  • Affiliations:

  • Graduate University of Chinese Academy of Sciences, Beijing, China and Key Lab of Intell. Info. Process., Inst. of Comput. Tech., Beijing, China;Harbin Institute of Technology, Harbin, China;Inst. of Comput. Tech., CAS, Beijing, China;Graduate University of Chinese Academy of Sciences, Beijing, China and Key Lab of Intell. Info. Process., Inst. of Comput. Tech., Beijing, China

  • Venue:
  • Proceedings of the 4th International Conference on Internet Multimedia Computing and Service
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

In order to make the users to access their interested news content conveniently, news analysis has been a hot research topic for a long time. However, most of the previous works only focus on news event detection, tracking, etc. Little attention has been paid to news report difference analysis and comparison. News is full of people's cognition of the world. Because of the different background, ideologies, the cognition of the people is different from one to another. Then, the angle of the news report is different accordingly. In this paper we propose a novel scheme to summarize and compare news report from different communities by using news pictures. Two challenges are addressed: similar news reports summary generation and different news reports summary generation. Firstly, news pictures from different sources are downloaded. Then, the bag-of-visual word features are extracted from these pictures to represent their content. After that, based on the similarity and dissimilarity of the pictures from different sources, collaborative ranking is adopted to rank the images to obtain the news similar and difference summaries. Experimental results on the selected news topics are promising and demonstrate that the proposed approach is effective.