Started in 2001, YALE has been developed by the Artificial Intelligence Unit of the University of Dortmund. YALE is a powerful data-mining package that in many ways is similar to Weka. It implements many Weka’s learning algorithms and filters. Like Weka, it is designed to be used in practical data mining problems and in research. It offers an alternative to the Experimenter and KnowledgeFlow interfaces of Weka.

Platforms Available

Like Weka, YALE is implemented entirely in Java so it is platform independent. There is no installation required, all one has to do is to unzip the software into a folder.

Where to get

Yale is an open-source software package that is released with a GPL license like Weka. Both the software and the source code can be obtained from its website:

http://www-ai.cs.uni-dortmund.de/SOFTWARE/YALE/index.html

Data Management

YALE has a built-in graphical data editor and can read CSV, ARFF, BibTex, C4.5, Dbase, arbitrary text formats. It can also bring in data directly from a database. For preprocessing of data, YALE incorporates many of Weka’s many preprocessing filters.

Data Mining Methods

Most of Weka’s data mining methods are implemented in YALE.

Tree Algorithms

All of Weka’s Tree Algorithms are implemented in YALE.

Visualization And Analysis Tools

YALE has many visualization features. One can visualize tree models as in Weka. More advanced displays are accessed via generic 2D and 3D plotters tools that can be fed any numerical data and then display it graphically.

Notes

This software package is very similar to Weka, and indeed a lot of Weka is implemented into YALE. This section of the documentation will be updated with a more indepth coverage of YALE.