Advance Hybrid RF-GBC-RFE Wrapper-Based Feature Selection Techniques for Prediction of Autistic Disorder

Abstract

Autistic disorder is a premature developmental ailments characterized by impaired societal interaction and persistent verbal exchange with stereotyped conduct. Detecting autistic ailments at an early stage is time consuming and very expensive. Machine learning classifiers play an imperative role in the early detection of autism spectrum disorders. The intention of this article is to make people aware of the early deduction of ASD in affected children. We provide a new hybrid technique to select the Feature-RF-GBC-RFE model in this work using the feature-based recursive feature elimination (RFE) ensemble of the Random Forest (RF) and the Gradient Boosting Classifier (GBC). Feature selection is a system that derives a subset of the perfect capabilities of a predictive modeling dataset. The feature in the ASD dataset is analysed and reduced by age category in this article. The hybrid RF-GBC-RFE feature selection technique, ML techniques such as Random Forest, Support Vector Machine, Gradient Boosting Classifier, and AdaBoost are used to study the reduced feature set. The model's overall performance can be categorized into precision and sensitivity metrics. A hybrid RF-GBC-RFE feature selection strategy is proposed in a unique way that improves data classification accuracy.