A neuroevolutionary method for knowledge space construction

In this paper we propose a novel method for the construction of knowledge spaces based on neuroevolution. The main advantage proposed approach is that it more suitable constructing large than other traditional data-driven methods. core idea if states are considered as neurons in neural network, optimal topology such network also space. To apply neuroevolutionary method, set analogies between and networks was established described paper. This evaluated comparison with minimized corrected inductive item tree analysis, de facto standard algorithm space construction, confirms assumptions.