Abstract
A first order regression algorithm capable of handling real-valued variables is introduced and some of its applications are presented. Regressional learning assumes real-valued (continuous) class and discrete or real-valued variables. The algorithm combines regressional learning with standard ILP concepts, such as first order concept description and background knowledge. IN particular, our goals were to develop a system which can: induce first-order logic concepts which incorporate real-valued variables, use background knowledge in intensional form, model dynamic systems (learn from time series), partition attribute space to subspaces in order to find a regressional submodel for each subspace separately, end handle noisy data. An outline of the algorithm and the results of the system's application in some artificial and real-world domains are presented. Real-world domains comprise modelling of the water behavior in a surge tank and modelling of the workpiece roughness in the steel grinding process. The results confirm the advantage of combining of various techniques and again confirm that the comprehensibility of the induced knowledge in crucial for successfull application of AI techniques.