Diagnosis of Behavioural Disorders Using Swarm Intelligent Metaheuristic Techniques

Behavioural disorders typically manifest in childhood or adolescence and can have a profound impact on an individual's social, academic, and personal functioning. Here, two primary datasets related to BDs is collected and analysed to determine the classification of these disorders. This study proposes a novel hybrid, Improved Particle Swarm Optimization using Grey Wolf Optimization (IPSOGWO) to find an optimal set of features for Anxiety and Disruptive Behaviour Disorder. The proposed hybrid method addresses Particle Swarm Optimization Algorithm’s challenges by using parallel swarms, integrating Grey Wolf Optimization for ranking, and introducing a dynamic, non-linear adjustment of the velocity parameter. These enhancements aim to improve the algorithm's ability to find optimal solutions effectively while avoiding premature convergence to local optima. Various performance metrics are evaluated and compared with eight different Swarm Intelligence (SI)-based metaheuristic techniques viz. Ant Colony Optimizer (ACO), Particle Swarm Optimization (PSO), Firefly Algorithm (FA), Crow Search (CS), Grey Wolf Optimization (GWO), Ant Lion Optimizer (ALO), Whale Optimization Algorithm (WOA), and Emperor Penguin Optimizer (EPO). As far as the classification rate is concerned, the (IPSOGWO) outperform individual SI techniques.