Treffer: Machine learning Models Compositions Builder.

The article presents the results of the development of a software product for constructing compositions of machine learning methods that gives an opportunity to solve the classification tasks in any scope of activities. The software tool supports a generalized procedure for solving classification tasks using machine learning methods, including preparing initial data, its preliminary analysis, building compositions of machine learning models, training them, presenting the characteristics of learning results, and classification of new objects. It is possible to choose different tools to solve the corresponding tasks at each step of research. The application architecture is flexible enough and allows to add new system components. Python Programming Language was used to build the software. The software tool can be used in areas where it is required to determine whether objects belong to a particular group and predict their behavior. A user-friendly interface enables business analysts and product managers without programming skills to conduct complex analysis. [ABSTRACT FROM AUTHOR]