A Comparison of Immune and Genetic Algorithms for Two Real-Life Tasks of Pattern Recognition

The field of GAs is well-established nowadays and it forms an important part of evolutionary computation and computational intelligence [1,2]. However, the computing capabilities of the natural immune system have only recently been appreciated within a new field of Artificial Immune Systems (AISs) [3,4]. The mathematical formalization of these capabilities [5] forms the basis of IC as a new computing approach that replicates the principles of information processing by proteins and immune networks [6]. Besides, IC models of immune networks [6,7,8] differ from those utilizing within AIS. Although AISs exploit some ideas of GAs, thus enhancing one another (see, e.g., [9,10]), no direct comparison between IC and GA has yet been available. This paper investigates more specifically this area, which has been explored before in a rather conceptual way.