Machine Learners

A number of different machine learning algorithms have been used in the my project, from the industry standard C4.5 to CaMML developed by researches in the computer science department at Monash University. The machine learners were used implemented with the use of Weka - developed at the University of Waikato, New Zealand and is available at www.cs.waikato.ac.nz/ml/weka. It contains Java implementations of a number of machine learners that were used to help understand the data that I had. Lucas Hope a PHD student at Monash University has incorporated some of the Machine Learners developed by other researches at Monash University for evaluation purposes. His software not only allows the user to run a machine learner on one machine learning algorithm, but allows for a number of the machine learners to be run at once and evaluated on the same testing and training data sets. This simplified the evaluation of the machine learners output significantly. I was able to write a wrapper, like Luke had previously done for his machine learners, to include the structure of a model that I developed with the aid of Netica (see my Bayesian Networks page) and test my models against the machine learners that learned the structure as well as parameterization from the data.