Abstract
An algorithm is presented for the induction of first-order logic concepts that incorporate real-valued variables. The algorithm enables use of background knowledge in intensional form, modelling of dynamic systems (learning from time series), partitioning of the attribute space to subspaces to find a regressional submodel for each subspace separately, and handling of noisy data. An outline of the algorithm and the results of its application in some artificial and real-world domains are presented. Real-world domains comprise modelling of the behavior of water in surge tank and modelling of the operator's behavior during the process of electrical discharge machining.