Prediction of Autism Spectrum Disorder Using Pheromone Updated Firefly Optimized Support Vector Machine

M. Premasundari and C. Yamini

Autism Spectrum Disorder (ASD) is the most common neurodevelopment disorder in children. The identification of autism is possible based on symptoms. One of the significant research issues is the diagnosis and classification of autism spectrum disorder. The aim of this research intends to predict the level of autism using support vector machine. Support vector machine (SVM) is the most widely used supervised learning technique for prediction or classification. The accuracy of the SVM classifier mainly depends on tuning the parameters of SVM. In this paper, the parameters of SVM are optimized by proposed Firefly algorithm to improve the efficiency of the classifier. The proposed method uses Pheromone Updated Firefly algorithm (PUFA) to optimize SVM parameters and it is applied for the prediction of autism levels. The experimental outcomes accomplished that the proposed method provided successful results to find the global optima when compared to other state-of-the-art techniques. The results specify that the proposed SVM-PUFA method can be acknowledged as a capable machine learning technique for accurate diagnosis of autism spectrum disorder.

Volume 11 | 04-Special Issue

Pages: 2129-2136