Detecting Harmful Web Documents Based on Web Document Analyses 


Vol. 12,  No. 5, pp. 683-688, Oct.  2005
10.3745/KIPSTD.2005.12.5.683


PDF
  Abstract

A huge amount of web documents, which are published on the Internet, provide to users not only helpful information but also harmful information such as pornography. In this paper we propose a method to detect the harmful web documents effectively. We first analyze harmful web documents, and extract factors to determine whether a given web document is harmful. Detail criteria are also described to assign a harmfulness score to each factor. Then the harmfulness score of a web document is computed by adding the harmfulness scores of all factors. If the harmfulness score of a web document is greater than a given threshold, the web document is detected as harmful. It is expected that this study could contribute to the protection of users from harmful web documents on the Internet.

  Statistics


  Cite this article

[IEEE Style]

K. H. Kim, J. M. Choi, J. H. Lee, "Detecting Harmful Web Documents Based on Web Document Analyses," The KIPS Transactions:PartD, vol. 12, no. 5, pp. 683-688, 2005. DOI: 10.3745/KIPSTD.2005.12.5.683.

[ACM Style]

Kwang Hyun Kim, Joung Mi Choi, and Joon Ho Lee. 2005. Detecting Harmful Web Documents Based on Web Document Analyses. The KIPS Transactions:PartD, 12, 5, (2005), 683-688. DOI: 10.3745/KIPSTD.2005.12.5.683.