Yale 3.0 released

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

Yale 3.0 released

Ingo Mierswa
Hello,

we would like to introduce the new version 3.0 of YALE (Yet
Another Learning Environment) to you:

    http://yale.cs.uni-dortmund.de

The new version includes a huge number of major improvements. Some of
them are listed below, the complete list of changes can be found at:

    http://sourceforge.net/project/shownotes.php?release_id=341383

YALE provides more than 200 operators for data mining and machine
learning and allows the design of complex process chains/trees. Of
course WEKA is also fully integrated.

YALE is a freely available open source software under the terms of the
GNU General Public License. Since YALE is entirely written in Java, it
runs on any major platform/operating system. You are welcome to use
it!


New features
------------

Several new features and operator were implemented, like:

  * The set of new operators include (excerpt):
      - different feature filters
      - multiple label handling
      - data generators
      - new performance criteria
      - several discretization schemes
      - transformation into value series
      - several parameter optimization operators
      - ARFFWriter
      - JViToPlotter (additional to build in plotters)

  * New Weka version 3.4.5 integrated

  * New LibSVM version 2.8 integrated

  * Improved data management and statistics. Yale can handle
    larger data sets now

  * Undo and Redo function

  * Some operators are deprecated now. Deprecated operators
    provide messages during application and validation

  * New plotter concept, introducing Yale color plotter,
    GnuPlotPlotter for 3D plots, scatter plots, and
    distribution plotter (histograms)

  * JViToPlotter provides a huge amount of 2D and 3D plots for
    high-dimensional data and models

  * Syntax highlighting

  * stratified crossvalidation

  * Several new selection and crossover schemes for EAs

  * Many operators can now pass through their input

  * and more


Bugfixes
--------

  * bug in unix start scripts
  * variance adaption in feature weighting
  * wrong conversion from Weka instances to Yale example sets
  * average handling of validation operators
  * strange plotting of some example sets
  * validation of experiments containing disabled operators


New plugin versions
-------------------

We released a new version of all YALE plugins, too:

   * Clustering plugin version 2.0

   * Value Series plugin version 2.0

   * Word Vector plugin version 2.0


Have fun and best regards,
Ingo Mierswa

--
Ingo Mierswa
Artificial Intelligence Unit
Department of Computer Science
University of Dortmund
47221 Dortmund, Germany
WWW: http://www-ai.cs.uni-dortmund.de
Mail: [hidden email]

_______________________________________________
Wekalist mailing list
[hidden email]
https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist