[PDF][PDF] Comparison of a heuristic method with a genetic algorithm for generation of compact rule based classifiers
SK Halgamuge, A Brichard, M Glesner - … of the 1995 ACM Symposium on …, 1995 - dl.acm.org
SK Halgamuge, A Brichard, M Glesner
Proceedings of the 1995 ACM Symposium on Applied Computing, 1995•dl.acm.orgRule based transparent classifiers can be generated by partitioning the input space into a
number of sub spaces. These systems can be considered as fuzzy classifiers assigning
membership functions to the partitions in each dimension. A flexible genetic algorithm based
method is compared with a heuristic method proposed iu this paper considering the
performance and running time. It is shown that for complex real world type of applications, a
preprocessing step with neural clustering methods reduces the running time of the genetic …
number of sub spaces. These systems can be considered as fuzzy classifiers assigning
membership functions to the partitions in each dimension. A flexible genetic algorithm based
method is compared with a heuristic method proposed iu this paper considering the
performance and running time. It is shown that for complex real world type of applications, a
preprocessing step with neural clustering methods reduces the running time of the genetic …
Abstract
Rule based transparent classifiers can be generated by partitioning the input space into a number of sub spaces. These systems can be considered as fuzzy classifiers assigning membership functions to the partitions in each dimension. A flexible genetic algorithm based method is compared with a heuristic method proposed iu this paper considering the performance and running time. It is shown that for complex real world type of applications, a preprocessing step with neural clustering methods reduces the running time of the genetic algorithm baaed method drastically.
ACM Digital Library