Weka: Practical machine learning tools and techniques with Java implementations

Abstract

Introduction The Waikato Environment for Knowledge Analysis<E-174> (Weka) is a comprehensive suite of Java class<E-188> libraries that implement many state-of-the-art<E-145> machine learning and data mining algorithms.<E-161> Weka is freely available on the World-Wide Web<E-189> and accompanies a new text on data mining [1]<E-196> which documents and fully explains all the<E-195> algorithms it contains. Applications written using<E-186> the Weka class libraries can be run on any<E-193> computer with a Web browsing capability; this<E-193> allows users to apply machine learning techniques<E-165> to their own data regardless of computer platform.<E-162> Tools are provided for pre-processing data, feeding<E-178> it into a variety of learning schemes, and analyzing<E-169> the resulting classifiers and their performance. An<E-196> important resource for navigating through Weka is<E-202> its on-line documentation, which is automatically<E-153> generated from the source.<E-77> The pri