Adaptive autism behavior prediction using improved binary whale optimization technique

Abstract

Summary Autism spectrum disorder (ASD), a neuro developmental disorder is a bottleneck to several clinical researchers owing to the data modularization, subjective analysis and shifts in the accurate prediction of the disorder among the sample population. Subjective clinical analysis suffers from lengthy procedure which is a time‐consuming process. The present research focuses on the prediction of ASD disorder using improved binary whale optimization that provides accuracy in the feature selection for the contribution towards the disorder and improves the accuracy in decision making of predicting the presence of disorder. The proposed technique is carried out in two steps: the acute features contributing to the disorder is selected using the binary whale optimization method and the optimal feature is subjected to that render decision of predicting the presence of ASD. The state‐of‐the‐art disorder dataset is tested with conventional techniques like particle swarm optimization (PSO), genetic algorithm (GA), particle swarm optimization with genetic algorithm (PSO‐GA), whale optimization method and binary whale optimization method. Based on the results, the improved binary whale optimization method is proposed and validates the effectiveness of the deciding the autism spectrum disorder.