Implementation and Analysis of the Parallel Genetic Rule and Classifier Construction Environment

This title is printed to order. This book may have been self-published. If so, we cannot guarantee the quality of the content. In the main most books will have gone through the editing process however some may not. We therefore suggest that you be aware of this before ordering this book. If in doubt check either the author or publisher’s details as we are unable to accept any returns unless they are faulty. Please contact us if you have any questions.

This paper discusses the Genetic Rule and Classifier Construction Environment (GRaCCE), which is an alternative to existing decision rule induction (DRI) algorithms. GRaCCE is a multi-phase algorithm which uses evolutionary search to mine classification rules from data. The current implementation uses a genetic algorithm based 0/1 search to reduce the number of features to a minimal set of features that make the most significant contributions to the classification of the input data set. This feature selection increases the efficiency of the rule induction algorithm that follows. However, feature selection is shown to account for more than 98 percent of the total execution time of GRaCCE on the tested data sets. The primary objective of this research effort is to improve the overall performance of GRaCCE through the application of parallel computing methods to the feature selection algorithm. The development and implementation of a parallel feature selection algorithm is presented. The experiments designed and used to test this parallel implementation are outlined followed by an analysis of the results. The results of this thesis effort show clearly that GRaCCE is improved through the use of parallel programming techniques.