Based on the cellular automata in artificial life, an artificial
Ants Sleeping Model (ASM) and an ant algorithm for cluster analysis
(A
^4
C) are presented. By simulating the swarm intelligence
of the real ant colonies, we use the ant agent to represent the data object. In
ASM, each ant has two states: sleeping state and active state. The ant's state
is controlled by a function of the ant's fitness to the environment it locates
and a probability for the ants to become active. The state of an ant is
determined only by its local information. By moving dynamically, the ants form
different subgroups adaptively, and consequently the data objects they
represent are clustered. Experimental results show that the
A
^4
C algorithm on ASM is significantly superior to other
clustering methods in terms of both speed and quality. It is adaptive, robust
and efficient.
