A negative selection algorithm for classification and reduction of the noise effect

摘要: Artificial Immune Systems (AIS) are a type of intelligent algorithm inspired by the principles and processes human immune system. In last decade, applications AIS have been studied in various fields. application change/anomaly detection, negative selection algorithms successfully applied. However, not appropriate for multi-class classification problems, because they do mechanism to minimize danger overfitting oversearching. this paper, we propose new overcome drawback extend area classification. The is named Negative Selection Classifier (ANSC). We investigate tolerance ANSC against noise, introduce method reduce effect noise into ANSC. accuracy data reduction compared with those from Recognition System (AIRS), which well known effective classifier AIS. results show that our useful problems effect.