The primary objective of this research is to explore the interplay of the relationship between depression severity, coursework performance, and ADHD in computer science students to build sound insight into the psychosocial aspects that influence both the academic and personal lives of students. This study is carried out by computer science students' survey (100), which consists of psychosocial factors like age, gender, GPA, annotations, depression status, completing academic tasks, presentation preferences, sleep rest, number of friends, and vulnerability to new encounters. Feature selection is carried out to distinguish vital features and to increase the power to predict, which involves utilization of the Stochastic Fractal Search (SFS) algorithm and its hybrid combined with the Whale Optimization Algorithm (WOA) algorithm and other variants. We have applied our selected features by creating a sequel to the classification machine learning models like Random Forest, Logistic Regression, K-Nearest Neighbor and others. The finding of the predictor showed that the bSFS-Guided WOA algorithm obtained the lowest average error, 0.131917, being regarded as the most efficient feature selection. In the classification models group, Random Forest came out first with the highest accuracy, 0.973449669, which implies its prominence in predicting the interdependence relationships. Discoveries highlighted psychological factors that affect student life and emphasized that mental wellness and study habits are essential for a student's academic success. The study suggests specific programs based on the findings and recommends thorough analysis to discover other factors that can be examined on a larger dataset.