A Study on Improving Tactical Strategies in Sports Games Using Optimization Algorithms

Sports are dynamic, unpredictable battles that require teams to develop strategic plans for making use of their advantages and adjust to changing conditions during play. In this study, we use optimization algorithms to improve the tactical methods which are used in sports games. Tactics are the intentional, premeditated moves, choices, and configurations made by groups or individuals to accomplish particular goals and outplay their rivals. The number of corresponding solutions addresses the problem of designing security strategies optimally using a game theory simulation. In this simulation, players choose to attack the structure based on certain unpredictable attack scenarios, while the attacker selects the way to defend the system by managing the available options for strategic solutions. The unpredictable nature of human performance is a barrier to the advancement of sports strategy tactics, since players may depart from pre-planned plans as a result of weariness, injuries, or emotional states. We suggested the self-regulated immune cat swarm optimization (SRICSO) strategy as a solution to these issues to improve tactical strategies in sports. A National Basketball Association (NBA) dataset was gathered for this study’s evaluation. The collected dataset underwent preprocessing using the minmax normalization technique, and feature extraction was accomplished through the use of principal component analysis. We compare our method’s precision (87%), accuracy (97%), F2-score (95%), and recall (92%) with more traditional methods such as the optimized convolutional neural network (OCNN), naïve Bayes (NB), and random forest (RF). According to the findings of an assessment conducted with a dataset from the NBA, the SRICSO approach was shown to have higher performance in terms of improving tactical strategies in sports.