Real-World Applications - Immunological Computation: Theory and Applications - page 172

Solution

Immune algorithms

Affinity measures

Representation

Immune entities

Application domain

Figure 7.1 Solving a problem using IC.

should be to decide which A IS will be better to generate a set of suitable entities that
can provide a good solution to the problem at hand. Figure 7.1 shows the necessary
steps to solve problems using an immunological approach.

7.2 Applications in Computer Security

Computer security seems to be analogous to the biological defense in many respects
(Forrest et al., 1996, 1997); thus we can learn a lesson from the immune system to
develop digital immunity. Majority of AIS works have been devoted to using some
immunological metaphor for developing digital defense systems (Aicklen et al., 2003;
Dasgupta, 1999; D'haeseleer et al., 1996). AISs used varied notions of data protection
and anomaly to provide a general-purpose protection system to augment current com-
puter security systems. h e security of computer systems depends on activities such as
detecting unauthorized use of computer facilities, maintaining the integrity of data
fi les, and preventing the spread of computer viruses. h is immunity-based system is
much more sophisticated.
Forrest et al. (1994) fi rst proposed the usage of negative selection (NS) in com-
puter security. h ey assumed the problem of protecting computer systems from
harmful viruses as an instance of the general problem of distinguishing “self ”
(legitimate users, uncorrupted data, etc.) from the dangerous “other” (unauthorized
users, viruses, and other malicious agents). h is method was intended to be comple-
mentary to the more traditional cryptographic and deterministic fi le-authentication
methods on the problem of computer virus detection.

7.2.1 Virus Detection

In this application, the NS algorithm (discussed in Section 4.2) was used to detect
changes in the protected data and program fi les. Experiments were performed