Immunity-Based Computational Models - Immunological Computation: Theory and Applications - page 45

Chapter 3
Immunity-Based
Computational Models
h e fi eld of immunological computation (IC) or artifi cial immune system (AIS)
has been evolving steadily (Dasgupta, 1999; Forrest et al., 1994; Tarakanov and
Dasgupta, 2000) since 1985. h ere has been an increasing interest in the devel-
opment of computational models inspired by several immunological principles
(Perelson and Oster, 1979; Percus et al., 1993). Some models intend to mimic the
abstract mechanisms in the biological immune system (BIS) to better understand
its natural processes and simulate its dynamic behavior in the presence of antigens
or pathogens; others, however, emphasize on designing artifacts—computational
algorithms, techniques using simplifi ed concepts (sometime obsolete) of various
immunological processes, and functionalities (Farmer et al., 1986; Hofmeyr and
Forrest, 2000; De Castro and Von Zuben, 2000; Stepney et al., 2004). Table 3.1
summarizes the mostly studied computational models of BIS, whereas the details
of these are described in Chapters 4 through 6. It shows the use of specifi c immu-
nological concepts in diff erent models and their intended applications (discussed
in Chapter 7).
Common terminologies that are used in most immune algorithms and their
corresponding terms used in machine learning are listed in Table 3.2.
h is chapter focuses on describing some common features that are used in most
immunity-based models. h ey use computational features like shape-space repre-
sentation, a nity measures, and immunity-based processes (Figure 3.1).
45