eHarmony / mallet

Bugfix fork of MAchine Learning for LanguagE Toolkit

Fork of Mallet, MAchine Learning for LanguagE Toolkit in java

Updated to last version: 2.0.7, 2011-9-22

Original mercurial repository's history kept using hg-git

Selected changes

• Improved documentation

• Pipes to get features from single tokens through transformations. See package cc.mallet.pipe.tsf.transform
  For example (TokenTransform) to convert to lower case or to convert to different morphologies, like U27k9 -> A11a1 (cap num num low num)