Quality Assessment of Wikipedia Content Using Topic Models

  • Lauro C. J. Santos CEFET-MG
  • Taís Christofani CEFET-MG
  • Ismael S. Silva CEFET-MG
  • Daniel H. Dalip CEFET-MG

Resumo


The web has become a large knowledge provider for society, allowing people to not just consume information but also produce it. Collaborative documents bring some significant advantages and decentralization, but they also raise questions concerning its quality. In this work, we explore the quality assessment on collaborative documents using these documents’ topics. The proposed approach improved in 3.2% the accuracy of quality assesment of Wikipedia content. Then, the main contribution in this paper is an analysis of how we can use topic modelling in order to improve quality prediction performance.
Publicado
29/10/2019
Como Citar

Selecione um Formato
SANTOS, Lauro C. J.; CHRISTOFANI, Taís; SILVA, Ismael S.; DALIP, Daniel H.. Quality Assessment of Wikipedia Content Using Topic Models. In: SIMPÓSIO BRASILEIRO DE SISTEMAS MULTIMÍDIA E WEB (WEBMEDIA) , 2019, Rio de Janeiro. Anais do XXV Simpósio Brasileiro de Multimídia e Web. Porto Alegre: Sociedade Brasileira de Computação, oct. 2019 . p. 249-252.