Estimating the parameters of the logistic regression model using the swam algorithm for the handball team of the College of Physical Education and Sports Sciences Al-Mustansiriyah University

This research investigated the biomechanical variables of movement analysis and its essentialcomponents for handball players at the College of Physical Education and Sports Sciences, Al-MustansiriyaUniversity, during a league match and university qualifiers. The data was analyzed using binary logisticregression, a mathematical model that defines the link between a dependent variable, which takes the valueof one when a goal is scored against the opposing team and zero when no goals are scored, and independentfactors.The bird swarm algorithm will be used in this research. It is one of the artificial intelligencealgorithms that rely on the intelligence of living flocks by monitoring their movements, such as birds, bees,cats, chickens, and many other swarm algorithms.The conclusions we reached from this study are as follows: When using logistic regression, we foundonly four explanatory variables that affect the dependent variable. They are detailed as follows: The twoexplanatory variables (The maximum height of the hip and Flight time until leaving the ball) and thedependent variable (shooting) have an inverse relationship and affect it. The two variables (Knee angle at themoment of thrust and The instantaneous speed of the ball) have a positive relationship with aiming and affectit.When we used the Bird Swarm algorithm, we found that all the explanatory variables included in thestudy had a significant effect on the dependent variable. The variables (Knee angle at the moment of thrust,Rising angle, Flight angle, The instantaneous speed of the ball, and the horizontal distance of theperformance) have a positive relationship, with the dependent variable (shooting). In contrast (The maximumheight of the hip and Flight time until leaving the ball) have an inverse relationship with the dependentvariable.Using the logistic model helps sports coaches and researchers to estimate and predict models,especially when the dependent variable takes values ​​(one or zero). In contrast, we noticed that the resultswere more accurate and objective when using the bird swarm algorithm. It further helps academics, thoseinterested in sports, and coaches benefit from these results.