In this paper, a novel negative selection algorithm for recognition problems was given. Compared with the traditional negative selection algorithm, a co-stimulation signal was added to start the detectors, which a key factor in immune response. Co-sti...
In this paper, a novel negative selection algorithm for recognition problems was given. Compared with the traditional negative selection algorithm, a co-stimulation signal was added to start the detectors, which a key factor in immune response. Co-stimulation signal was calculated by the techniques of the statistics and the sliding window, which not only reduced time complexity of algorithm but also improved accuracy of the algorithm. Entropy was adopted to evaluate the density of detectors for optimizing the coverage of nonself area. Experiment results proved high accuracy and efficiency of the proposed algorithm.