timbl 6.4.2-1 (amd64 binary) in ubuntu precise

 Memory-Based Learning (MBL) is a machine-learning method applicable to a wide
 range of tasks in Natural Language Processing (NLP).
 .
 The Tilburg Memory Based Learner, TiMBL, is a tool for NLP research, and for
 many other domains where classification tasks are learned from examples. It
 is an efficient implementation of k-nearest neighbor classifier.
 .
 TiMBL's features are:
  * Fast, decision-tree-based implementation of k-nearest neighbor
 classification;
  * Implementations of IB1 and IB2, IGTree, TRIBL, and TRIBL2 algorithms;
  * Similarity metrics: Overlap, MVDM, Jeffrey Divergence, Dot product, Cosine;
  * Feature weighting metrics: information gain, gain ratio, chi squared,
 shared variance;
  * Distance weighting metrics: inverse, inverse linear, exponential decay;
  * Extensive verbosity options to inspect nearest neighbor sets;
  * Server functionality and extensive API;
  * Fast leave-one-out testing and internal cross-validation;
  * and Handles user-defined example weighting.
 .
 TiMBL is a product of the ILK Research Group (Tilburg University, The
 Netherlands) and the CLiPS Research Centre (University of Antwerp, Belgium).
 .
 If you do scientific research in NLP, timbl will likely be of use to you.