Title :
A Novel Immune Algorithm for Supervised Classification Problem
Author_Institution :
Modern Educ. Center (MEC), HeNan Radio & Telev. Univ., Zhengzhou, China
Abstract :
This article presents a novel immune algorithm for a solution of supervised classification problem .The algorithm is based on the risk model, the use of dangerous and hazardous signal mechanism, the risk by assessing the Antigen to the signal classification; and use of antibody-Antigen interactions learning mechanisms to make antibodies have strong populations of adaptive learning capacity. Simulation results show that the algorithm have good classification results and learning performance compared with other traditional algorithm.
Keywords :
artificial immune systems; learning (artificial intelligence); pattern classification; adaptive learning capacity; antibody-antigen interactions learning mechanisms; dangerous signal mechanism; hazardous signal mechanism; immune algorithm; learning performance; risk model; signal classification; supervised classification problem; Classification algorithms; Cloning; Complexity theory; Data models; Immune system; Joining processes; Training;
Conference_Titel :
Control, Automation and Systems Engineering (CASE), 2011 International Conference on
Conference_Location :
Singapore
Print_ISBN :
978-1-4577-0859-6
DOI :
10.1109/ICCASE.2011.5997726