Analysis of Behavioral Motivation in a Self-Management Model of Physical Activity

Abstract

In this paper, we preprocess data in self-management of physical exercise according to different dataset attributes, perform attribute coding commonly used features binary, and use data standardization to scale physical exercise data attributes to mean values. Supervised information of physical exercise can be achieved to minimize the training error. Regression classification based on elastic networks is performed to obtain important physical activity self-management feature sets. Indicators in the database reflecting the impact of physical exercise on the body, and accordingly the set of attributes of the database is noted as attribute values. The metabolic data under the elastic network algorithm is 0.6, and the number of iterations is sustained at a peak of 570 times, which is conducive to the development of physical exercise and self-management outside the physical education classroom.