Machine learning for the classification of motor competence with wearable technology in schoolchildren

Abstract

Personal health can be determined by adequate physical activity Motor competence is an important aspect to help this and must be carried out from school days. The objective of this study was to assess motor competence with wearable technology, generate the percentiles of the evaluation metrics, and classify motor performance using machine learning techniques in primary and secondary school children. For this, smart bands were used as wearable technologies for data capture during the evaluation of motor skills tests in schoolchildren from educational centers. The CRISP-DM methodology was followed, and the data set consisted of 485 schoolchildren between 7 and 18 years of age. As a result of the application of machine learning algorithms, the best precision was achieved with the decision tree with 96.97% in the classification of motor performance in these students. It is concluded that using smart bands allows better data capture and processing precision to classify motor skills tests in schoolchildren better and can be used by interested persons.