☼ 林永忠 ☼ > HTML-Similarity-0.2.0 > HTML::Similarity

Download:
HTML-Similarity-0.2.0.tar.gz

Dependencies

Annotate this POD

CPAN RT

Open  1
View/Report Bugs
Source  

NAME ^

HTML::Similarity - Calculate the structural similarity between two HTML documents

SYNOPSIS ^

  use HTML::Similarity;

  my $hs = new HTML::Similarity;

  my $a = "<html><body></body></html>";
  my $b = "<html><body><h1>HOMEPAGE</h1><h2>Details</h2></body></html>";

  my $score = $hs->calculate_similarity($a, $b);
  print "Similarity: $score\n";

DESCRIPTION ^

This module is a small and handy tool to calculate structural similarity between any two HTML documents. The underlying algorithm is quite simple and straight-forward. It serializes two HTML tree to two arrays containing node's tag names and finds the longest common sequence between the two serialized arrays.

The similarity is measured with the formula (2 * LCS' length) / (treeA's length + treeB's length).

Structural similarity can be useful for web page classification and clustering.

PREREQUISITE ^

HTML::DOM, Algorithm::LCS

COPYRIGHT ^

Copyright (c) 2011 Yung-chung Lin.

This library is free software; you can redistribute it and/or modify it under the same terms as Perl itself.

syntax highlighting: