Application of Inductive Modeling Principles to Solve the Double Clustering Problems
Volodymyr Osypenko, Valentyna Osypenko · 2021 IEEE 16th International Conference on Computer Sciences and Information Technologies (CSIT) · 2021
An algorithm for solving a so-called Double Clustering problem is considered. The criterion used in the algorithm is to minimize the distances between the centers of clusters obtained regardless of the goal features of objects and their input features, which can be considered analogous to the criterion of consistency of inductive modeling of complex systems. An example presented in the article shows that underused criterion conditions, the algorithm proposed here can be used to choose the informative features for the regression model, for example, as well.