Data classification by Fuzzy Ant-Miner
In this paper we propose an extension of classification algorithm based on ant colony algorithms to handle continuous valued attributes using the concepts of fuzzy logic.
The ant colony algorithms transform continuous attributes into nominal attributes by creating clenched discrete intervals. This may lead to false predictions of the target attribute, especially if the attribute value history is close to the borders of discretization.
Continuous attributes are discretized on the fly into fuzzy partitions that will be used to develop an algorithm called Fuzzy Ant-Miner. Fuzzy rules are generated by using the concept of fuzzy entropy and fuzzy fitness of a rule.
Keywords: Fuzzy Ant Miner, fuzzy entropy, fuzzy fitness, discretization on the fly, classification.
Download Full-Text
ABOUT THE AUTHORS
Mohamed Hamlich
Mohamed Hamlich is affiliated with Computer science Lab. of FST, Mohammadia, Morocco. His research interests include Data Classification, knowledge extraction and artificial intelligence. He affords scientific advice to a group of research in cardiology at CHU, University Hassan II, Casablanca, Morocco.
Mohammed Ramdani
Mohammed Ramdani is affiliated with Computer science Lab. of FST, Mohammadia, Morocco. His research interests include Fuzzy logic, Data Classification, knowledge extraction and artificial intelligence.
Mohamed Hamlich
Mohamed Hamlich is affiliated with Computer science Lab. of FST, Mohammadia, Morocco. His research interests include Data Classification, knowledge extraction and artificial intelligence. He affords scientific advice to a group of research in cardiology at CHU, University Hassan II, Casablanca, Morocco.
Mohammed Ramdani
Mohammed Ramdani is affiliated with Computer science Lab. of FST, Mohammadia, Morocco. His research interests include Fuzzy logic, Data Classification, knowledge extraction and artificial intelligence.
