Improving diet through a recommendation system using physical activity data and healthy diet indexes of female futsal players

Abstract

Sports data analysis can revolutionize how coaches and athletes train, leading to enhanced skills and improved team outcomes by providing valuable insights into performance metrics, enabling personalized training programs, and fostering a data-driven approach to decision-making. Sports data analysis has evolved in tandem with the increasing availability of data and the widespread adoption of data-driven practices in sports. Futsal stands out as one of the most challenging team sports to analyze, particularly in the case of female futsal, which has received limited research attention. The high-paced nature of the game, the smaller playing area, and the emphasis on close ball control necessitate a specialized approach to data collection and analysis. This situation presents a significant opportunity for in-depth exploration. By recognizing a gap in utilizing nutrition and physical activity data for female futsal players, researchers embarked on the journey to design and develop a recommendation system based on diet and training data. The study enlisted 14 talented female futsal players, and data was collected using an advanced LPS (Local Positioning System) device. Despite having limited player information, researchers successfully addressed the well-known “cold start” challenge. They created a content-based filtering recommendation system that accurately predicts the caloric expenditure of futsal players, achieving an impressive determination coefficient of 0.94. This innovative system has the potential to revolutionize the training methods of female futsal players, paving the way for advancements in sports data analysis and opportunities to enhance the visibility of women’s futsal on a broader stage.