International Journal of Engineering and Technology

The rate of Children influenced with extreme autism spectrum disorder (ASD) is expanding rapidly. The extremely autistic children confront the issues in comprehension aptitudes, thinking capacity, learning and related abilities. The troubles that the extremely autistic children experience are various. The autistic Children frequently encounter trouble in learning aptitudes, interaction with others and dealing with the social abilities. The most fundamental part of the paper is to examine and to propose the best optimal and ideal features to surmount the learning obstacles and to accelerate the learning capacity of autistic children. The technique proposed comprises of three stages. The First stage identifies the Feature Extraction. The Second stage indicates the Feature Selection. Third stage adopts the classifier methodology. The Principal Component Analysis act as an evaluator in the Feature Extraction. Particle Swarm Optimization is utilized as a Feature Selection. The Particle Swarm Optimization assesses the parameter optimization to retrieve the most effective and best subset of features. The successful best subsets of features are fed into the Support Vector Machine classifier. A definitive result thus emerge show that the proposed technique acquires a higher accuracy with good exactness.