Autism Spectrum Disorder Detection using Navie Bayes Tree Technique

Autism Spectrum Disorder (ASD) impacts the social and communicative abilities of individuals of all ages. Autism screening is a time-consuming and complicated procedure, making it difficult to discover autistic individuals. Recently, machine learning techniques have been applied to streamline the screening process and rapidly discover ASD. However, it is necessary to increase the accuracy of predictions using the limited data available. This research provides a novel approach for detecting ASD using the Nave Bayes (NB) Tree. The proposed method generates a decision tree from a data set using the naive Bayes categorizer at the nodes to assess whether or not a person has autism. The performance of the suggested technique is compared to that of the most commonly used classification algorithms, including Nave Bayes, Support Vector Machine (SVM), and Multilayer Perceptron (MLP). The results of the experiments show that the NB Tree-based technique is better than the current classification methods in terms of classification accuracy, Root Mean Squared Error (RMSE), and precision.