Classification Algorithm Based On Ais

Currently, AIS has been a new branch in the field of computing intelligence, which has already shown a powerful capability in the information process and problem solutions, such as data mining, machine learning, automatic control and fault diagnosis. The researches based on AIS in the field of data mining were put emphases on data clustering, data concentrating and classification.This paper draws on the inspiration of artificial immune system and brings forward a new classification algorithm based on AIS. The algorithm use the binary string as the problem presentation and take the attribute of the problem space as the gene. The special combination of the attribute makes up of the chromosome. The characteristic of the algorithm is using the negative selection mechanism as the filter. At first the rule set comes from the samples, and then the negative operators are used to eliminate a part of rules which could make the immune recognition. After that the rule set is evolved by the genetic algorithm with the negative operators. At last the final rule set is got. We use a group of data from the UCI database for experiments. The results of the experiments show that the algorithm has a good classification capability.