Data classification based on a modified fuzzy artificial immune network

Based on fuzzy theory and artificial immune network, a new data classification method, named fuzzy artificial immune network classification (FAINC), is put forward. In order to improve the convergence rate, fuzzy C mean clustering algorithm is employed to provide vaccines (initial population) for artificial immune network. The operators such as clonal selection, network compression, immune maturation and immune memory are explored. According to expanding and compressing the population of the network, a stable antibody network can be acquired. Namely, the antibody network represents the concentrated training data to construct the classificator. The simulation experiments of University of California, Irvine (UCI) data sets validate the higher classification accuracy, data enrichment, stability and reliability of FAINC than aiNet.