Simulation and Modeling of Expert System for Prediction of Autism using Naïve NLP Method

Abstract

A developmental disorder that impacts behavior and communication is called autism spectrum disorder (ASD). The main objective of this venture is to establish an expert system which applies natural language processing (NLP) and machine learning techniques to make ASD predictions. The research examines computational approaches while recommending an original method for early autism detection which depends on natural language processing models. The expert system performs analysis of voice and textual information to recognize ASD-related linguistic patterns while boosting the early diagnosis precision. Laboratory results demonstrate that the Proposed Model produces results better than SVM and Naïve Bayes models by achieving 99.8% accuracy for all tests. Naïve Bayes stands as less accurate than the SVM because it delivers 94.2% performance however Naïve Bayes only reaches 90.8% accuracy. The Proposed Model demonstrates its capacity for delivering exceptional classification outputs which makes it an effective and dependable solution for predictive duties. The technology advances diagnostic precision through machine learning algorithms that allow early intervention thus leading to better ASD support and management systems.