I've seen many implementations of machine learning techniques in Python, C++ , Java and Matlab, but never much for Perl. The most perl has to offer is a interface to packages written for many different languages. There are thus no native implementations found on cpan.

The same applies for many AI applications. Many ai programs with apis rarely advocate perl.

I'm curious to know why this turned out so. As far as I know perl is a pretty powerful language that makes it quite useful for NLP.

asked Sep 28 '11 at 17:05

mugetsu's gravatar image

mugetsu
233212431


2 Answers:

Probably because of the lack of numpy equivalent that allows to do efficient and high level numerical processing and linear algebra which is often required to implement machine learning algorithms.

answered Sep 28 '11 at 17:20

ogrisel's gravatar image

ogrisel
498995591

1

There's Perl Data Language, but it has never had much success.

(Sep 28 '11 at 18:24) Gael Varoquaux

good find there Gael, never knew about PDL. Wonder why there hasn't been much success in that case.

(Sep 28 '11 at 19:00) mugetsu

as much as i love perl, the code is unreadable. not well suited for a collaborative software project. especially when compared to python.

answered Sep 28 '11 at 19:30

downer's gravatar image

downer
54891720

Your answer
toggle preview

powered by OSQA

User submitted content is under Creative Commons: Attribution - Share Alike; Other things copyright (C) 2010, MetaOptimize LLC.