The purpose of this research was the paradigm model of teaching physical education lessons based on machine learning using the foundation's data approach.This research is based on the philosophy (paradigm)of pragmatism, qualitative in terms of method, and data-driven in terms of strategy.The research population includes experts in the research topic, including faculty members of physical education and sports sciences faculties across the country who have at least the rank of associate professor or professor and have more than 10 years of teaching experience.The sampling method in the qualitative part of the research was conducted using two targeted interview tools and the snowball technique, and 13 people were interviewed to theoretical saturation. The validity of the research, according to Guba and Lincoln(2000), included four criteria of validity, transferability, reliability, and confirmability of the validation study.The Lavish method was used to evaluate the reliability of the interview protocol. In this study, the opinions of 11 experts were obtained for validation using the Lavish method, which was ultimately calculated based on the results of 5 item scores less than 0.6, so these items were removed from the model and a 65-component model was extracted. Based on the findings, the results of a 13-factor model with 65 factors were presented in the form of a data-driven model with the Strauss and Corbin approach for implementing physical education lessons based on machine learning. Using machine learning algorithms, patterns and algorithms can be developed that can predict the performance and progress of each student/student or athlete.