Detection of Autism Spectrum Disorder using Select Classification Algorithms

Abstract

Autism spectrum disorder (ASD) is a neurological and developmental disorder that often appears in infancy and impacts various developmental processes. This study highlights the use of multiple machine-learning algorithms to accurately diagnose autism spectrum disorder (ASD). It explores the application of classification algorithms, namely Support Vector Machine (SVM), k-Nearest Neighbours (KNN), Logistic Regression (LR), Neural Network (NN), and Random Forest (RF), for ASD detection. The main objective of the project is to assess the precision with which these algorithms can identify persons with ASD and those without. A comparative analysis measures the accuracy of Logistic Regression, KNN, Neural Network, Random Forest, and SVM. The project's overarching objective is to attain high accuracy rates and elevated levels of precision and recall metrics, pivotal for robust and dependable ASD detection. The remarkable performance of the Support Vector Machine (SVM) algorithm is of particular significance, as it achieved an unprecedented accuracy rate of 99.9%. This result underscores SVM's potential as an effective tool for precise ASD identification. The project's findings underline the synergy between advanced computational methods and medical diagnosis, illustrating the capacity of machine learning to aid clinicians and diagnosticians in early ASD detection. In conclusion, this project contributes to advancing ASD diagnosis by strategically selecting and comparing classification algorithms. The exceptional accuracy achieved by SVM signifies a pivotal stride forward in the quest for accurate and dependable ASD detection methods. This study exemplifies the potential of interdisciplinary collaboration between technology and healthcare, focusing on achieving high accuracy and comprehensive metrics for the improved identification and understanding of autism spectrum disorder in patients