Weka 3.4.13 (book) and 3.5.8 (developer)

New versions of Weka are available for download from the Weka homepage:

  • Weka 3.4.13 - the book version available as ZIP, Win32 installer, Win32 installer incl. JRE 1.4.2_13 and Mac OS X application.

  • Weka 3.5.8 - the developer version available as ZIP, with Win32 installer, Win32 installer incl. JRE 1.5.0_16 and Mac OS X application.

Both versions contain a significant number of bugfixes, it is recommended to upgrade to the new versions. The documentation on the WekaDoc Wiki on Sourceforge.net will be updated in the near future.

Weka homepage:
http://www.cs.waikato.ac.nz/~ml/weka/

Note:
It might take a while before Sourceforge.net has propagated all the files to its mirrors.

What's new in the developer version?

Some highlights

  • New classifiers: Bayesian logistic regression, discriminitive multinomial naive Bayes for text classification, functional trees and decision table-naive Bayes hybrid.
  • SubsetByExpression filter.
  • Latent semantic analysis.
  • Improved output in Logistic, NaiveBayes, EM and SimpleKMeans.
  • Plugin support for the KnowledgeFlow.
  • Ability to execute knowledge flows outside of the GUI.
  • Output predictions for a run of cross-validation and percentage split on the command line.
  • Instance weights can now be specified in a standard ARFF file.
  • Cost sensitive attribute selection.

New classes:
weka.attributeSelection.ASEvaluation:
weka.attributeSelection.CostSensitiveAttributeEval
weka.attributeSelection.CostSensitiveSubsetEval
weka.attributeSelection.FilteredAttributeEval
weka.attributeSelection.FilteredSubsetEval
weka.attributeSelection.LatentSemanticAnalysis

weka.classifiers.Classifier:
weka.classifiers.bayes.BayesianLogisticRegression
weka.classifiers.bayes.DMNBtext
weka.classifiers.rules.DTNB
weka.classifiers.trees.FT

weka.core.DistanceFunction:
weka.core.EditDistance

weka.clusterers.Clusterer:
weka.clusterers.CLOPE
weka.clusterers.sIB

weka.classifiers.functions.supportVector.Kernel:
weka.classifiers.functions.supportVector.PrecomputedKernelMatrixKernel

weka.experiment.SplitEvaluator:
weka.experiment.DensityBasedClustererSplitEvaluator

weka.associations.Associator:
weka.associations.HotSpot

weka.filters.Filter:
weka.filters.unsupervised.instance.SubsetByExpression

As usual, for a complete list of changes refer to
the changelogs.

Cheers,
The Weka Team

Posted by Mark Hall 2008-07-16