BayesTH-MCRDR Algorithm for Automatic Classification of Web Document

  • Woo-Chul Cho
  • Debbie Richards
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3339)

Abstract

Nowadays, automated Web document classification is considered as an important method to manage and process an enormous amount of Web documents in digital forms that are extensive and constantly increasing. Recently, document classification has been addressed with various classified techniques such as naïve Bayesian, TFIDF (Term Frequency Inverse Document Frequency), FCA (Formal Concept Analysis) and MCRDR (Multiple Classification Ripple Down Rules). We suggest the BayesTH-MCRDR algorithm for useful new Web document classification in this paper. We offer a composite algorithm that combines a naïve Bayesian algorithm using Threshold and the MCRDR algorithm. The prominent feature of the BayesTH-MCRDR algorithm is optimisation of the initial relationship between keywords before final assignment to a category in order to get higher document classification accuracy. We also present the system we have developed in order to demonstrate and compare a number of classification techniques.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Woo-Chul Cho
    • 1
  • Debbie Richards
    • 1
  1. 1.Department of ComputingMacquarie UniversitySydneyAustralia

Personalised recommendations