Search

In this paper we suggest new modifications of the multiobjective genetic algorithm - NSGA-II, which are aimed to improve quality classification ratings in the SVM classifier development. The proposed modifications of genetic operators in NSGA-II, autotuning algorithm in particularly allows to minimize the dependence of the NSGA-II result from the parameters' values which are set manually. The experimental results obtained on the basis of the model and real datasets of loan scoring, medical diagnostics, psychognosis, etc. confirm the efficiency of the proposed algorithms and the practicability of further research in this field. [ABSTRACT FROM AUTHOR]

In this paper the questions of the choice of the training data sequence for the forecasting model on the base of the strictly binary trees which is usually applied to time series with a short length of actual part (about 15 – 30 elements) have been considered. It is offered to use the principles of the attractors’ forming in the presence of the long time series that will allow creating the training data sequence more reasonably. The reviewed examples confirm the efficiency of the attractors’ use in sense of minimization of the affinity indicator of the forecasting model, and also the forecasting errors on 1 – 5 steps forward. Besides, the minimization of time expenditures on development of the forecasting model is provided. [ABSTRACT FROM AUTHOR]