Indeed, Text Mining is a very large subject... A Roadmap to Text Mining and Web Mining is one starting point, even if only available through the "Wayback Machine". Text Mining, Web Mining, Information Retrieval and Extraction from the WWW is another one.

Once you have a good general overview of the subject, you have determined that Perl is the best choice of language for what you intend to do, and have some specific Perl questions in mind, don't hesitate to come back and ask. :-)

HTH,