An Anomaly Detection-Based Classification System

Haiyu Hou, G. Dozier
2006 IEEE International Conference on Evolutionary Computation  
In this paper, we describe the construction of a classification system based on an anomaly detection system the employs constraint-based detectors, which can be generated using a genetic algorithm. The performance of the classification system was evaluated using the two benchmark datasets including Wisconsin Breast Cancer Dataset and Fisher's Iris Dataset. A typical AIS generates detectors using the Negative Selection Algorithm [6]. Firstly, candidate detectors are randomly
doi:10.1109/cec.2006.1688584 dblp:conf/cec/HouD06 fatcat:b3w2gzau3zcepbhovt5lqpxeaq