Использование гематологических показателей для прогноза результатов соревнований спортсменов пловцов на основе вероятностных оценок

Abstract

Purpose: The article discusses probabilistic approaches to the analysis of hematological indicators of athletes in swimming, their use for assessing the functional state in order to predict the success of performances in competitions and the creation of a computer system for assessing the prediction of successful performances based on the most significant hematological signs. Methodology of the work: On the basis of probabilistic approaches, a method for assessing the influence of hematological parameters on the determination of the functional state (in order to predict the success of performances) of athletes in competitions is proposed. Two assessments of the significance of indicators are considered: an integral assessment and an assessment of the influence of the values of each feature on the forecast result. The integral significance assessment is designed to select the features that most strongly affect the prediction result, and the assessment of the influence of the values of each feature allows you to determine the impact of the values of each indicator on the success prediction and serves to explain the prediction result. Results: The use of estimates of the integral significance of signs made it possible to reduce the number of hematological blood parameters from 18 to 8 in order to develop a computer system for evaluating the prediction of the success of athletes in competitions. A technique for converting continuous indicators into a discrete scale with three gradations based on the criterion of the maximum integral significance of the indicator is proposed. An assessment of the influence of each feature gradation on the probability of a successful forecast is given. A computer system for predicting the success of the performance of athletes has been developed. A technique for assessing the quality of the conclusions of the system for predicting the success of the performance of athletes based on the Monte-Carlo methods is proposed. Scope of application of the results: The results of the work are of applied importance for the construction of computer systems for assessing the functional state and predicting the success of athletes’ performances in order to improve the quality of the selection of candidates both in team disciplines and in individual competitions of athletes. Conclusions: The use of assessments of the integral significance of signs made it possible to reduce the number of hematological indicators for the development of a system for assessing the prediction of the success of athletes’ performance in competitions. The results of statistical testing of the methodology for assessing the quality of the conclusions of the system for predicting the success of the performance of athletes on Monte-Carlo approaches showed a fairly high accuracy (about 93%) of the prediction system on the available volume of actual data, which allows us to conclude that it is working and the possibility of further development.