DMelt:AI/Machine Learning algorithms
Summary of Machine Learning in DataMelt
DataMelt includes a reach variety of various machine learning algorithms implemented over the past 20 years by the Java community. DataMelt includes easy-to-use GUI-based tools, as well as Java classes that can easily be embedded inside an Java code, or using with scripting languages, such as Jython, Groovy and JRuby.
You can learn about different methods and algorithms inside DataMelt using DataMelt Search
Encog Workbench v3.4 by Heaton Research using the menu "Tools - -Neural Networks: Encog". This program can be run using the GUI mode as described above, or calling the Java classes directly inside the Java code, or scripting macro languages. This section of the DataMelt manual describes how to call Encog classes directly inside analysis code.
Joone Workbench, or Java Object Oriented Neural Engine" can run as "Tools-Neural Networks: Joone". This program can be run using the GUI mode as described above, or calling the Java classes directly inside the Java code, or scripting macro languages. This section on Weka usage describes this.
Weka neural network algorithms are also included as Java libraries. You can also use Weka in the GUI mode using the menu "Tools - Neural Networks: Weka". Note that Weka scans only jar files inside the directories "user", "weka" and "math". Other DataMelt Java libraries are not visible for Weka.
If you use Jython/Groovy and other scripting engines, you can call appropriate Java classes in your Java code or scripting macros.
Convolutional Neural Networks
Convolutional Neural Networks are included from several Java libraries. They are described in this section of the manual. No GUI Workbench is available.
Bayesian Neural Networks
I've discovered several implementations of Bayesian Neural Networks inside DataMelt API. This section  has a description of some of them.