An Immuno-inspired Distributed Artificial Classification System

With the advent of distributed systems such as the Internet of Things (IoT), there has been a surge in the amount of data being generated by the computational nodes. Classification of these datasets remains a challenging problem, especially in distributed scenarios where the data is not situated within a single node. This paper describes a novel immuno-inspired Distributed Artificial Classification System (iDACS) to solve this problem within a distributed network of nodes. It uses a combination of mechanisms from the Clonal Selection theory, the Immune Network theory, and the Danger theory, to realize a system that can solve the classification problem in a distributed manner. The results obtained from experiments performed on real networks when compared with five other standard classification models revealed that the proposed system was more suited to distributed scenarios.