Improved Traditional Fitness Model by Applying Big Data Analysis

Abstract

This study elaborated on the importance of fitness in the contemporary environment, put forward the problems in traditional fitness, and conducted a series of discussions according to the questions. It conducted an in-depth analysis of fitness data utilising appropriate data analysis techniques to explore the relationship between different fitness data. Moreover, this study explores the processes and tools needed for analysis and explains the difficulties and resistance that may be encountered in future research. The literature section provides a detailed discussion on muscle gain and weight loss in fitness, the elaboration of big data frameworks, and machine learning methods that may be applied in this field. However, the regression models were only conducted on calorie burning for weight loss due to the lack of suitable muscle data. The optimal Mean Absolute Error and coefficient of determination were obtained as 8.307 and 0.967. The final section also concludes the process and results of this study and puts forward the shortcomings and the direction for future improvement.