Experts use behavioral and cognitive assessments to classify ADHD subtypes because they have not identified dependable neurophysiological biomarkers yet. This research evaluates the forecasting capabilities of combined EEG data alongside cognitive screening tests as well as ADHD symptom records for subtype identification in ADHD patients. The analysis through PCA and the KNN classification demonstrated self-reported symptoms to be the most accurate predictor of ADHD subtypes with a recorded accuracy rate of 73.68%. The combination of EEG recordings and behavioral data proved useful for predicting participant sex at 87.4% accuracy though they showed restricted capability in diagnosing ADHD subtypes. EEG signals prove insufficient for diagnosing ADHD subtypes independently yet they demonstrate excellent potential as brain signatures for separating physiological patterns related to male and female participants. The research demonstrates that ADHD diagnostic improvements require multicomponent strategies wherein neurophysiological information delivers essential value for developing person-centered therapeutic plans. Future research should work to develop superior machine learning models using larger datasets alongside deep learning methods for improving prediction accuracy. Functional connectivity studies and long-term observational approaches can lead to better understanding of ADHD neurobiological foundations which would advance the development of specific medical tools and intervention strategies.