What's new or improved in Weka 3.7.3

Classifiers

REPTree has been improved to consume less memory. Now has an option to use a Leplace-like correction for probability estimates computed at the leaves.

InputMappedClassifier is a new meta classifier that addresses differences between the structure of data used to train its base classifier and that of incoming data to be classified. It builds a mapping between features and values of nominal features.

Package manager

The package manager now checks for new packages at the repository. Now allows "unofficial" packages to be directly installed from a URL or file in the GUI.

New packages introduced in 3.7.3:

  • WekaExcel - a new package that provides loaders and savers for Microsoft Excel spreadsheets.
  • WekaODF - a new package that provides loaders and savers for spreadsheets in the open document format.
  • classAssociationRules - a new package that provides the ability to use association rule learning algorithms as classifiers. Includes an implementation of the CBA algorithm.
  • kfPMMLClassifierScoring - a new package that provides a component for the Knowledge Flow for scoring using PMML classifiers.
  • multiInstanceLearning - (version 1.0.1) now has an implementation of the quick diverse density learner (QuickDDIterative).

Package changes: