IDGSC: An Immunity-Based Approach for Dynamic Grid Service Classification

Service classification is not only an indispensable method for manually browsing and managing service repositories, but can in fact be used for simple but efficient forms of service annotation and semi-automated discovery, complementary aid for automatic selection. In order to classify services in a service taxonomy dynamically, a new service classification model based on immune theory is proposed in this paper. The concepts of "self", "nonself" and "filter" are given first. Then the model is described in detail. Finally, a simulation experiment is performed, and the important parameters of the model are analyzed. The experimental result shows that the new model possess better adaptability and stability.