Towards improving wikipedia as an image-rich encyclopaedia through analyzing appropriateness of images for an article

  • Authors:
  • Xinpeng Zhang;Yasuhito Asano;Masatoshi Yoshikawa

  • Affiliations:
  • Kyoto University, Kyoto, Japan;Kyoto University, Kyoto, Japan;Kyoto University, Kyoto, Japan

  • Venue:
  • APWeb'11 Proceedings of the 13th Asia-Pacific web conference on Web technologies and applications
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

In Wikipedia, knowledge related to an object is gathered into a single article that is mainly composed of text and images. We observed that some Wikipedia images are inappropriate for addition to some Wikipedia articles. In this paper, we propose a RCT model for analyzing the appropriateness of images for Wikipedia articles. The model assumes that an image appearing in article s indicates relations related to s. The model then examines the appropriateness of an image through analyzing relations indicated by the image. We also propose a system which examining Wikipedia images using the RCT model. The system also searches images on the Web which could serve as references for users to add appropriate images or text to edit high-quality and image-richWikipedia articles. Finally, we confirm through experiments that the RCT model could examine the appropriateness of Wikipedia images to a satisfactory degree.