G. Stiglic, P. Kokol (Slovenia), and L. Lhotska (Czech Republic)
Artificial intelligence, machine learning, multi-agent
systems, software fault prediction.
In this paper we propose a new method inspired by a
multi-agent based system that was initially used for
identification of significant genes in microarray
databases. Gene subset selection is a common problem in
the filed of bioinformatics. If we regard the software
measurements values of a software module as a genome
of that module, and the real world dynamic characteristic
of that module as its phenotype (i. e. failures as a disease
symptoms) we can borrow the established bioinformatics
methods in the manner first to predict the module
behaviour and second to data mine the relations between
metrics and failures.