Only in the purpose. While Lingua::Ident supports six languages and only implements one method (the bigram method), Lingua::Identification supports a few more languages (already twelve, allowing the possibility of learning new ones in a much easier way) and implements a bunch of other methods (users can choose which one to use or let the module choose for them, depending on things as the number of words to identify, the length of the text, etc; this is very important). Lingua::Identification also allows for other things, as identifying the language of a file whether it is a pdf, html, ps, etc.
Also, Lingua::Ident needs to learn the languages each time it is used... the same does not hapen with Lingua::Identification, as all that work as been done in advance. Besides, Lingua::Ident doesn't appear to be maintained anymore... José Castro Citando "Sean M. Burke" <[EMAIL PROTECTED]>: At 03:45 PM 2003-06-16 +0200, Perl Authors Upload Server wrote: > modid: Lingua::Identification > DSLIP: bdpfg > description: Perl extension for language identification > userid: COG (José Alves de Castro) Does this relate to http://search.cpan.org/author/MPIOTR/Lingua-Ident-1.4/Ident.pm at all? -- Sean M. Burke http://search.cpan.org/~sburke/ __________________________________________________________ O email preferido dos portugueses agora com 100Mb de espaço e acesso gratuito à Internet http://www.portugalmail.com/gold/
