Efficient Diagnosis of Autism with Optimized Machine Learning Models: An Experimental Analysis on Genetic and Personal Characteristic Datasets

Early diagnosis of autism is extremely beneficial for patients. Traditional approaches have been unable to diagnose in a fast and accurate way; rather, there are multiple factors that can be related identifying the disorder. The gene expression (GE) individuals may one these factors, addition personal behavioral characteristics (PBC). Machine learning (ML) based on PBC GE data analytics emphasizes need develop prediction models. quality relies accuracy ML model. To improve prediction, optimized feature selection algorithms applied solve high dimensionality problem datasets used. Comparing different methods using bio-inspired over types allow most model identified. Therefore, this paper, we investigated enhancing classification process spectrum disorder 16 proposed models (GWO-NB, GWO-SVM, GWO-KNN, GWO-DT, FPA-NB, FPA-KNN, FPA-SVM, FPA-DT, BA-NB, BA-SVM, BA-KNN, BA-DT, ABC-NB, ABC-SVM, ABV-KNN, ABC-DT). Four namely, Gray Wolf Optimization (GWO), Flower Pollination Algorithm (FPA), Bat Algorithms (BA), Artificial Bee Colony (ABC), were employed optimizing wrapper method order select informative features increase Five evaluation metrics used evaluate performance models: accuracy, F1 score, precision, recall, area under curve (AUC). obtained results demonstrated achieved good as expected, with accuracies 99.66% 99.34% by GWO-SVM datasets, respectively.