Prediction of Attention Deficit Hyperactivity Disorder (ADHD) in Regional Locations Using Novel Support Vector Machine Algorithm Over the Naive Bayes Algorithm for Better Accuracy

The study's focus is to enhance the precision and accuracy of novel ways for anticipating attention-deficit/hyperactivity disorder at local sites utilizing the machine learning Support Vector Machine algorithm. Materials and Methods: The Novel Support Vector Machine classifier is applied to an ADHD dataset containing 1395 records, 119 characteristics, 104 samples A framework comparing SVM and Naive Bayes classifiers was presented and created for the purpose of predicting the regional distribution of ADHD in the medical area. The classifiers were evaluated and their precision and accuracy were noted. Results: On the dataset provided, the Naive Bayes classifier predicts the regional location of ADHD disorder with an accuracy of 89.12%, while the Support Vector Machine classifier predicts the same with a 96.75% accuracy. There is a statistically significant difference in the groupings between these two. has a 95% confidence interval of
$(\mathrm{p}=0.010;$
p0.05). Conclusion: In comparison to Naive Bayes, Novel Support Vector Machine appears to perform better and produce more accurate and reasonable results.