Project

Autism Detection via Questionnaire (ML – Logistic Regression)

Home

"Developed a machine learning model using Logistic Regression to support early autism screening with questionnaire-based data. Achieved ~86.9% accuracy with high recall, minimizing false negatives, and prepared the model for user-friendly deployment via Streamlit."

2025

Overview

A machine learning model for early autism screening using questionnaire-based data, incorporating demographic features and clinical indicators. Logistic Regression was identified as the most effective model, achieving ~86.9% accuracy with high recall—ideal for preliminary screening, especially in resource-limited settings.

Techniques Used

Results