Free Decision Tree Tools 2015
Gambit is a bit different, and more sophisticated than usual. It is an open-source collection of tools for doing computation in game theory, and by implication decision trees. With Gambit, you can build, analyze, and explore game models.
SilverDecisions currently provides a browser based interface for manual construction of decision trees. However by right clicking in the browser it can be installed on a local computer. Resulting trees can be saved as images (png) or in a native dtf format. It is developed at Decision Support and Analysis Division, Warsaw School of Economics and is written in C#.
Simple Decision Tree is an Excel Add-in created by Thomas Seyller. The Add-in is released under the terms of GPL v3 with additional permissions.
The five decision tree tools listed below are essentially data mining tools, and are not manual decision tree tools often used in decision support. Three of these tools are stand-alone and the other two come as part of an extensive data mining environment.
This decision tree tool makes use of genetic algorithms to evolve binary decision trees. It allows users to set the characteristics of the resulting decision tree and can provide a set of different decision trees that match the solution space. The free version restricts the number of generations and the maximum size of the initial population. The commercial version costs 200 Euro.
Decision trees are one of the methods supported by KNIME, the free data mining workbench. Although not as powerful as some of the stand-alone tools, the user interface is friendly and most business problems can be addressed with it.
Decision trees are implemented within RapidMiner, the free data mining environment, and although not as powerful as some of the stand-alone decision tree tools, nonetheless it is capable of addressing many business problems.
Extends the classical decision tree (new splitting criteria, non-greedy search etc) and works as a classifier or probability estimator. Versions available for UNIX and Windows.
Yet another Decision Tree builder (YaDT) employs entropy based tree construction algorithms and runs on Windows and Linux. It implements a more efficient version of C4.5 and computes the same trees, but with a performance gain of up to five times.
A free introductory book “Decision Tree Primer” by Rafael Olivas can be downloaded here.