Exploring time management proficiency with a classification approach to assessing students' temporal skills.

Time is undoubtedly one of humanity's most essential and priceless resources, and its very nature is irreversible. Poor time regulation skills, especially among students, can swiftly lead to poor academic accomplishment. For the improvement of their conditions, people with time management issues usually require interventions like occupational therapy. The current research utilized well-chosen input data from reliable sources. The students' performance was forecasted by the base model, i.e. light gradient boost classification (LGBC). For enhancing outcome accuracy, two metaheuristic algorithms, i.e. Northern Goshawk optimization (NGO) and sand cat swarm optimization (SCSO) were used. The hybrid schemes developed from this process were named ligh gradient boost classification with northern Goshawk optimization (LGNG) and light gradient boost classification with sand cat swarm optimization (LGSC) in the present study. Moreover, SHapley Additive exPlanations (SHAP) sensitivity appraisal was utilized to determine the input factors most significantly influencing the outcome. The LGNG model stands out among the schemes' classification outcomes by having a marvelous 95.2% accuracy compared to other schemes. Notably, it records the highest classification accuracy and the minimum error rate, which implies that it is stable and confident. Therefore, this approach possesses tremendous potential in the prediction of student performance and can lead institutions in the direction of improvement in management plans. [ABSTRACT FROM AUTHOR]