Advertisement

Truth Discovery Based on Crowdsourcing

  • Chen Ye
  • Hongzhi Wang
  • Hong Gao
  • Jianzhong Li
  • Hui Xie
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8485)

Abstract

Truth discovery is an important component of data cleaning and information integration. However, in the absence of knowledge, some truth could not be found from databases themselves. A possible solution is to involve crowds to find all the truth with the knowledge of crowds. In this paper, we propose a truth discovery framework based on active learning model with crowdsourcing. First, we give the basic voting algorithm BVote . Then we present the simple crowding-based truth discovery framework STDA based on BVote. Experimental results show that the STDA framework for truth discovery has improved significantly in accuracy with minimal efforts of workers.

Keywords

truth discovery crowdsourcing active learning 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Yin, X., Han, J., Yu, P.S.: Truth discovery with multiple conflicting information providers on the web. In: KDD 2007 (2007)Google Scholar
  2. 2.
    Galland, A., Abiteboul, S., Marian, A., Senellart, P.: Corroborating information from disagreeing views. WSDM 2010 (2010)Google Scholar
  3. 3.
    Dong, X.L., Berti-Equille, L., Hu, Y., Srivastava, D.: Global detection of complex copying relationships between sources. In: VLDB 2010 (2010)Google Scholar
  4. 4.
    Howe, J.: The rise of crowdsourcing. Wired Magazine 14(6), 1–4 (2006)MathSciNetGoogle Scholar
  5. 5.
    Cohn, D.A., Ghahramani, Z., Jordan, M.I.: Active learning with statistical models. arXiv preprint cs/9603104 (1996)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Chen Ye
    • 1
  • Hongzhi Wang
    • 1
  • Hong Gao
    • 1
  • Jianzhong Li
    • 1
  • Hui Xie
    • 1
  1. 1.Harbin Institute of TechnologyHarbinChina

Personalised recommendations