LAD-WEKA
This is the Web page of LAD-WEKA, a Java implementation
of the Logical Analysis of Data (LAD) classification algorithm
within the WEKA package.
About LAD:
Logical Analysis of Data (LAD) is a
rule-based machine learning algorithm based on ideas from Optimization and
Boolean Function Theory. The LAD methodology was originally conceived by
Peter L. Hammer, from Rutgers University, and has been
described and developed in a number of papers since the late 80's. It has
also been applied to classification problems arising in areas such as
Medicine, Economics, and Bioinformatics. A list with representative
publications about LAD will be made available here shortly.
About WEKA:
WEKA is an open-source environment for data analysis and machine learning, developed by the Machine Learning group of the University of Waikato, New Zealand. WEKA is the de facto standard for comparing machine learning algorithms and the tool of choice for many data mining and machine learning practitioners. The WEKA package provides access to a range of classification algorithms, including state-of-the-art implementations of SVM, Neural Networks, and Random Forests.
Goals
The goal of the LAD-WEKA project is to provide a reference implementation of LAD, which is free, portable, and does not depend on third-party software (such as linear and integer programming solvers). The fact that LAD-WEKA is implemented as a WEKA Classifier allows one to easily run experiments comparing LAD with other classification algorithms available in the WEKA package.
The current version of LAD-WEKA deals exclusively with two-class classification problems containing numerical or binary data (see the To-do list page for an outline of upcoming developments).
The current version of LAD-WEKA deals exclusively with two-class classification problems containing numerical or binary data (see the To-do list page for an outline of upcoming developments).
Development
LAD-WEKA was developed by Vaux S. D.
Gomes and Tibérius O. Bonates and funded by
CNPq, the Brazilian Council for Scientific and
Technological Development. It is currently
available as a stand-alone runnable file (a documented version of the
source code might be publicly available in the near future). To run the
application, the user must have the Java Runtime Environment (version 1.6
or later) installed in his/her computer. Check the Tutorial
page for
information on installing and running WEKA on your system.