Paper Title
SWARMPSYCHNET: A HYBRID SWARM INTELLIGENCE AND MACHINE LEARNING FRAMEWORK FOR DETECTING PSYCHOLOGICAL DISORDER

Abstract
Mental illness disorders like mood, anxiety, and sleep disorders are a growing concern, but diagnosis tends to remain based on clinical subjective assessments, which may lead to inaccuracies. The primary challenge addressed in this research lies in the imperative to have efficient, data-oriented, and automated diagnostic methods that can increase the accuracy and reliability of mental disorder detection. For addressing these issues, this research proposes Swarm PsychNet which is a combination of Swarm Intelligence (SI) feature selection schemes and machine learning classifiers to enhance classification accuracy. Two metaheuristic schemes, namely Ant Lion Optimization (ALO) and Grey Wolf Optimizer (GWO), have been implemented for feature selection, which lowered data dimensions and determined important mental indicators. For experimentation, we have taken the Psychological Assessment Dataset from Kaggle, which contains 9,504 data sets and 11 characteristics. It has been taken after preprocessing, that is, after normalization and label encoding. Optimized feature sets have been tried out RF and SVM classifiers. It has been found that GWO-RF demonstrated best performance, which attained an accuracy level of 99.53%, outperforming other settings in terms of precision, recall, and balanced accuracy. It has been indicated that the research has a strong capability to enhance interpretability of a set of models and increase diagnostic accuracy. Keywords - Psychological Disorder, Swarm Intelligence, Grey Wolf Optimizer (GWO), Ant Lion Optimization (ALO), Machine Learning, Artificial Intelligence (AI)