Abstract
We have developed a rule learning system that employed
Grammar Based Genetic Programming for knowledge
discovery from databases. A grammar is used as a
template for the rule format. Grammar Based Genetic
Programming can guide the evolution and the discovery
of meaningful rules. The technique Token competition is
used to achieve the learning of multiple rules
simultaneously. We have applied the approach to real
life medical databases. The rules discovered can
provide insight into and allow better understanding of
the medical domains.
Users
Please
log in to take part in the discussion (add own reviews or comments).