• Version
• Download 9
• File Size 395.73 KB
• File Count 1
• Create Date 24 July 2025
• Last Updated 24 July 2025

Autism Prediction Using Machine Learning

PINNAMRAJU T S PRIYA, ANNEPU VAMSI

HOD, Assistant Professor, MCA Final Semester, Master of Computer Applications, Sanketika Vidya Parishad

Engineering College, Vishakhapatnam, Andhra Pradesh, India

ABSTRACT

Autism Spectrum Disorder (ASD) is a complex neurodevelopmental condition that affects communication, social interaction, and behavior. Early identification of ASD is essential for timely intervention and improved developmental outcomes. This project aims to build an efficient and accessible predictive model using machine learning, specifically Logistic Regression, to identify individuals at risk of autism. The dataset used includes behavioral screening scores (A1–A10) along with demographic and medical features such as age, gender, ethnicity, jaundice history, and family history of autism. The data is preprocessed by handling missing values and encoding categorical variables using Label Encoding. To address class imbalance in the dataset, Random Over-Sampling is applied, which improves the model’s ability to generalize across both autistic and non-autistic classes. Comprehensive data analysis is carried out through statistical summaries and visualizations, including pairplots, heatmaps, count plots, and histograms. The model is trained and evaluated using metrics such as accuracy, classification report, and confusion matrix, achieving strong predictive performance. To make the solution practical and user-friendly, a Gradio-based web interface is developed, allowing users to input responses through sliders and dropdowns to receive real-time ASD predictions. This project demonstrates the potential of machine learning in supporting early diagnosis of autism and contributes to more accessible, fast, and reliable healthcare screening tools.

INDEX TERMS: Autism Spectrum Disorder (ASD), Machine Learning, XGBoost, Random Forest Classifier, Preprocessing, Medical Diagnosis, Feature Engineering, Behavioral Data, Classification Algorithm, Python, Data Visualization, ASD Detection, Model Evaluation, Early Disease Detection, Gradio Interface.