Learning (k,l)-contextual tree languages for information extraction

  • Authors:
  • Stefan Raeymaekers;Maurice Bruynooghe;Jan Van den Bussche

  • Affiliations:
  • Dept. of Computer Science, K.U.Leuven, Leuven;Dept. of Computer Science, K.U.Leuven, Leuven;Dept. Theoretical Computer Science, Universiteit Hasselt, Diepenbeek

  • Venue:
  • ECML'05 Proceedings of the 16th European conference on Machine Learning
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper introduces a novel method for learning a wrapper for extraction of text nodes from web pages based upon (k,l)-contextual tree languages. It also introduces a method to learn good values of k and l based on a few positive and negative examples. Finally, it describes how the algorithm can be integrated in a tool for information extraction.