Role of Morphometric Characteristics in Predicting Children's 20-Meter Sprint Performance: A Machine Learning Study

Thursday, 18 July 2024, 06:53

This study explores how morphometric features impact 20-meter sprint performance in children using machine learning algorithms. 130 male and 152 female volunteers aged 6-11 years were included. Different experiments were conducted to determine optimal ML techniques for prediction, emphasizing feature selection and dimensionality reduction. Key results show that correlation-based feature selection, including age, height, and circumferences, led to accurate predictions with a low Mean Squared Error of 0.012. Overall, this research provides valuable insights into identifying relevant predictors for sprint performance in children.

Nature — Role of Morphometric Characteristics in Predicting Children's 20-Meter Sprint Performance: A Machine Learning Study

Study Overview

This study examines morphometric features' impact on children's 20-meter sprint performance using machine learning techniques.

Participants

130 males and 152 females aged 6-11 years were involved in the study.

Experiments Conducted

Initial training on entire feature space.
Feature selection based on correlation analysis.
Application of Principal Component Analysis for dimensionality reduction.

Key Findings: Correlation-based feature selection yielded accurate predictions with minimal error (MSE 0.012).

This article was prepared using information from open sources in accordance with the principles of Ethical Policy. The editorial team is not responsible for absolute accuracy, as it relies on data from the sources referenced.

Study Overview

Participants

Experiments Conducted

Related posts