
Lingua::EN::Semtags::Engine - extract semantic tags (semtags) from English text

use Lingua::EN::Semtags::Engine;
my $engine = Lingua::EN::Semtags::Engine->new;
my @semtags = $engine->semtags("your blog post title");

Lingua::EN::Semtags::Engine uses Lingua::EN::Tagger and WordNet::QueryData to extract semantic tags (semtags) from English text. Semtags are words which reflect the semantic essence of a piece of text (similar to topic keywords).
Lingua::EN::Semtags::Engine was designed and developed to solve a particular problem I was facing.
Problem: a user is processing blog post titles and needs to programmatically determine the posts' semantic context.
Solution: the user feeds a blog post title to Lingua::EN::Semtags::Engine and gets back a set of semtags which can be used for further processing (e.g., web searches).
Example: a blog post title like "BBtv: Graffiti Research Lab, the movie" (boingboing.net, Xeni Jardin, April 24, 2008 8:00 AM) would produce the following semtags: DECORATION WORKPLACE SHOW.
Please note that the module makes the following assumptions when attempting to extract semtags:
PHRASE_FRAME_SIZE (set to 3) tokens;MIN_ISAS (set to 3) levels deep;SEMTAG_ISA_INDEX (set to 1, starts with 0) level of the hierarchy.Calls sentence($string), gets back a populated instance of Lingua::EN::Semtags::Sentence, iterates over its Lingua::EN::Semtags::LangUnits, populates and returns an array of their semtags.
Returns an instance of Lingua::EN::Semtags::Sentence, populates it with Lingua::EN::Semtags::LangUnit objects which represent meaningful language units.
Returns the Lingua::EN::Tagger instance used by the engine.
Returns/sets the verbose mode.
Returns the WordNet::QueryData instance used by the engine.

Lingua::EN::Tagger, WordNet::QueryData, http://code.google.com/p/lingua-en-semtags-engine

Igor Myroshnichenko <igorm@cpan.org>
Copyright (c) 2008, All Rights Reserved.
This software is free software and may be redistributed and/or modified under the same terms as Perl itself.