Learning from nowhere

We extend the Fuzzy Inference System (FIS) paradigm to the case where the universe of discourse is hidden to the learning algorithm. Hence the training set is constituted by a set of fuzzy attributes in whose correspondence some consequents are observed. The scenario is further complicated by the fact that the outputs are evaluated exactly in terms of the same fuzzy sets in a recursive way. The whole works arose from everyday life problems faced by the European Project Social&Smart in the aim of optimally regulating household appliances’ runs. We afford it with a two-phase procedure that is reminiscent of the distal learning in neurocontrol. A web service is available where the reader may check the efficiency of the assessed procedure.