Learning Fuzzy Rule Based Classifier in High Performance Computing Environment

An approach to estimate the number of rules by spectral analysis of the training dataset has been recently proposed (1). This work presents an analysis of such a method in high performance computing environment. Two approaches for parallel implementation of the method were studied considering the structure selection genetic algorithm and the spectral decomposition. The results show that both approaches have allowed to reduce considerably the overall processing time.

[1]  Eyke Hüllermeier,et al.  Fuzzy methods in machine learning and data mining: Status and prospects , 2005, Fuzzy Sets Syst..

[2]  Fan Chung,et al.  Spectral Graph Theory , 1996 .

[3]  Alexandre Evsukoff,et al.  Learning Fuzzy Rule Based Classifier with Rule Weights Optimization and Structure Selection by a Genetic Algorithm , 2007, 2007 IEEE International Fuzzy Systems Conference.

[4]  Hisao Ishibuchi,et al.  Rule weight specification in fuzzy rule-based classification systems , 2005, IEEE Transactions on Fuzzy Systems.

[5]  R. D. Santos,et al.  Comparing Two Parallel Genetic Algorithms For The Inverse Problem Associated To The Cardiac Bidomain Equations , 2007 .

[6]  George Karypis,et al.  Introduction to Parallel Computing , 1994 .

[7]  Francesco Masulli,et al.  A survey of kernel and spectral methods for clustering , 2008, Pattern Recognit..

[8]  Erick Cantú-Paz,et al.  A Survey of Parallel Genetic Algorithms , 2000 .

[9]  Anthony Widjaja,et al.  Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond , 2003, IEEE Transactions on Neural Networks.

[10]  Hisao Ishibuchi,et al.  Effect of rule weights in fuzzy rule-based classification systems , 2001, IEEE Trans. Fuzzy Syst..

[11]  Nelson F. F. Ebecken,et al.  Design of interpretable fuzzy rule-based classifiers using spectral analysis with structure and parameters optimization , 2009, Fuzzy Sets Syst..

[12]  Vicente Hernández,et al.  SLEPc: A scalable and flexible toolkit for the solution of eigenvalue problems , 2005, TOMS.