A danger theory inspired artificial immune algorithm for on-line supervised two-class classification problem

Self-nonself discrimination has long been the fundamental model of modern theoretical immunology. Based on this principle, some effective and efficient artificial immune algorithms have been proposed and applied to a wide range of engineering applications. Over the last few years, a new model called ''danger theory'' has been developed to challenge the classical self-nonself model. In this paper, a novel immune algorithm inspired by danger theory is proposed for solving on-line supervised two-class classification problems. The general framework of the proposed algorithm is described, and several essential issues related to the learning process are also discussed. Experiments based on both artificial data sets and real-world problems are carried out to visualize the learning process, as well as to evaluate the classification performance of our method. It is shown empirically by the experimental results that the proposed algorithm exhibits competitive classification accuracy and generalization capability.