Early Prediction of Attention Deficit Hyperactivity Disorder among Children's Using Support Vector Machine over the Linear Regression Algorithm for Better Accuracy

Abstract

This research work is to strengthen the precision and accuracy of forecasting factors impacting sex ratio in novel ADHD using machine learning algorithm (SVM) in comparison to linear regression method (LR). Materials and Methods: The SVM application of a classifier to a novel ADHD dataset with 1395 records, 119 attributes, 104 participants, two groups, and 80 percent g-power as well as kid patients with Novel ADHD were gathered from a number of websites, along with the most recent study results and 0.05 percent threshold, mean, standard deviation, and 95% confidence interval. A new novel architecture for predicting sex ratio in healthcare sector comparing SVM and LR classifiers was proposed and developed. The classifiers' precision and accuracy were measured, appraised, and reported. Results: The LR provides 81.12% in predicting the factors involved in Novel Attention Deficit Hyperactivity Disorder disease with various dataset, while the SVM classifier predicts the same at a rate of 92.54%, 0.05 percent threshold, 95% confidence interval, mean, and standard deviation.