A Gravitational-Double Layer Extreme Learning Machine and its Application in Powerlifting Analysis

Abstract

Powerlifting is a strength sport that is quite popular in the world. Powerlifters have their power levels varied at different ages and body weights, and their power levels are closely related to their performance. Therefore, studying the impact of age and weight on the performance of powerlifters is an important work. The traditional method relies mainly on artificial experience to judge the performance, and often does not get the desired results. In recent years, machine learning has developed rapidly, and applying machine learning in sports is a very interesting topic. This study is based on a new machine learning algorithm to construct a prediction model for the best performance of powerlifters. We propose a double-layer extreme learning machine based on affine transformation and two-layer extreme learning machine theory (AF-DELM). Then use a dynamic weight-gravitational search algorithm to improve the AF-DELM networks. The results show that the algorithm can better predict the performance and provide an effective predictive aid for the powerlifting competition.

Keywords

• Gravitational search algorithm
• gravitational-double layer extreme learning machine
• powerlifting performance
• prediction model