Towards a self regulating local network neighbourhood artificial immune system for data clustering

The theory of idiotopic lymphocyte networks in the natural immune system inspired the modelling of network based artificial immune systems (AIS). Many of these network based AIS models establish network links between the artificial lymphocytes (ALCs) whenever the measured Euclidean distance between the ALCs are below a certain network threshold. The linked ALCs represent an artificial lymphocyte network. Graaff and Engelbrecht introduced the Local Network Neighbourhood AIS (LNNAIS) [2]. The interpretation of the network theory is the main difference between LNNAIS and existing network based AIS models. The LNNAIS uses the concept of an artificial lymphocyte neighbourhood to determine network links between ALCs [A.J. Graaf and A.P. Engelbrecht, 2007]. The purpose of this paper is to highlight the drawbacks of the proposed LNNAIS model and to address these drawbacks with some enhancements, improving LNNAIS towards a self regulating AIS.