| This work constitutes a serious attempt and a real life application of one of machine learning and discovery's approaches, namely rough sets theory, in the field of biological control of weeds.; Different concepts of learning and discovery were surveyed and presented. The theory of rough set's different tools and techniques were also presented as new and alternative approaches in the rapidly growing fields of learning and discovery. The field of biological control of leafy spurge weed by a beetle called Aphthona nigriscutis was selected as a practical application of the theory of rough sets.; Not only have the obtained results confirmed the findings of respective researchers, but also suggested important data factors that researchers were not aware of. A set of factors was selected as the most important in evaluating new sites. Different combinations of factors, including the most important ones, were presented as alternatives that can be used to evaluate the appropriateness of sites before new releases are made.; Rough sets theory holds great promise for the badly needed automation of knowledge engineering process and provides a complementary data analysis for statistical methods. (Abstract shortened by UMI.)... |
