Applying machine learning on data using Java Machine Learning (Java-ML) library
Java Machine Learning (Java-ML) library is a collection of standard machine learning algorithms. Unlike Weka, the library does not have any GUI because it is primarily aimed at software developers. A particularly advantageous feature of Java-ML is that it has a common interface for each type of algorithm, and therefore, implementation of the algorithms is fairly easy and straightforward. The support for the library is another key feature of it since the source codes are well documented and hence extendable, and there are plenty of code samples and tutorials for all sorts of machine learning tasks that can be accomplished using the library. The website http://java-ml.sourceforge.net/ has all the details regarding the library.
In this recipe, we will use this library to do the following tasks:
- Dataset import and export
- Clustering and evaluation
- Classification
- Cross-validation and held-out testing
- Feature scoring
- Feature...
