An Open Source Pattern Recognition Toolbox for MATLAB

摘要

Pattern recognition and machine learning are becoming integral parts ofalgorithms in a wide range of applications. Different algorithms and approachesfor machine learning include different tradeoffs between performance andcomputation, so during algorithm development it is often necessary to explore avariety of different approaches to a given task. A toolbox with a unifiedframework across multiple pattern recognition techniques enables algorithmdevelopers the ability to rapidly evaluate different choices prior todeployment. MATLAB is a widely used environment for algorithm development andprototyping, and although several MATLAB toolboxes for pattern recognition arecurrently available these are either incomplete, expensive, or restrictivelylicensed. In this work we describe a MATLAB toolbox for pattern recognition andmachine learning known as the PRT (Pattern Recognition Toolbox), licensed underthe permissive MIT license. The PRT includes many popular techniques for datapreprocessing, supervised learning, clustering, regression and featureselection, as well as a methodology for combining these components using asimple, uniform syntax. The resulting algorithms can be evaluated usingcross-validation and a variety of scoring metrics to ensure robust performancewhen the algorithm is deployed. This paper presents an overview of the PRT aswell as an example of usage on Fisher's Iris dataset.