Autism spectrum disorder detection using machine learning.

Autism Spectrum Disorder also called as autism is a developmental illness with no complete treatment that is marked by issues with speech and communication. Autism is a complex, difficult and permanent developmental disorder that is characterized by issues with repetitive behavior, nonverbal communication and lack of focus. We have proposed a methodology in this paper that will help inthe early diagnosis of ASD. We have used a dataset of 1986 patients and applied 27 features that is available on the Kaggle. Then the artificial intelligence model is prepared with the dataset obtained from Kaggle and the comparison of the result was done with the KNN,Random Forest, Decision Tree, Gradient Boosting Classifier. The accuracy obtained by the machine learning classifier was 98.17% and when it was compared with other functional classifiers such as KNN, RF, DT, SVM, Gradient Boosting, Naïve Bayes (accuracy lies between 68-93%) it was found that proposed machine learning model achieved the highestaccuracy. Hence, we can say that the proposed model provides an effective way of Autism prediction. The result of the study confirms that our proposed machinelearning model obtained the best accuracy rate amongall the different classifiers used by different authors topredict Autism. [ABSTRACT FROM AUTHOR]