Information Technology Reference
In-Depth Information
A sonogram is plotted with the frequency components and power spectral density
values sequenced on the timeline. Time is on the x-axis, while frequency is on the
y-axis and gray value of the display represents the corresponding power spectral
density.
3 AIRS Classifier Algorithm
AIRS is a resource limited supervised learning algorithm inspired from immune
metaphors. In this algorithm, the used immune mechanisms are resource competition,
clonal selection, affinity maturation and memory cell formation. The feature vectors
presented for training and test are named as Antigens while the system units are called
as B cells. Similar B cells are represented with Artificial Recognition Balls (ARBs)
and these ARBs compete with each other for a fixed resource number. This provides
ARBs, which have higher affinities to the training Antigen to improve. The memory
cells formed after the whole training Antigens were presented are used to classify test
Antigens. The algorithm is composed of four main stages, which are initialization,
memory cell identification and ARB generation, competition for resources and
development of a candidate memory cell, and memory cell introduction. Table 1
summarizes the mapping between the immune system and AIRS.
Table 1.
Mapping between the Immune System and AIRS
Immune System
AIRS
Antibody
Feature Vector
Recognition Ball
Combination of feature vector and vector class
Shape-Space
Type and possible values of the data vector
Clonal Expansion
Reproduction of ARBs that are well matched antigens
Antigens
Training data
Affinity Maturation
Random mutation of ARB and removal of the least
stimulated ARBs
Immune Memory
Memory set of mutated ARBs
Metadynamics
Continual removal and creation of ARBs and memory
cells
