IDEAL: A Machine Learning Framework for the Automated Detection of Early Academic Learning Disabilities

With the increasing need for early identi- fication of learning disorders among children, there is a growing demand for intelligent, scalable, and child- friendly diagnostic systems. Learning disabilities such as dyslexia, ADHD, and related cognitive challenges often go undetected due to reliance on time-consuming manual assessments and subjective evaluations. The proposed system engages children through adaptive tasks to assess cognitive abilities such as memory, attention, logical reasoning, and language comprehension, reducing anx- iety while ensuring accurate data collection. The system incorporates an intelligent framework that combines in- teractive assessment techniques with data-driven analysis to support early identification of learning disorders. A machine learning–based backend utilizes classifiers like Random Forest, Extra Trees, and Logistic Regression to analyze performance data and detect potential learning difficulties, with comparative evaluation to determine the most effective model. Additionally, a dedicated dashboard enables educators and specialists to visualize results and track progress over time, providing a scalable and efficient solution for continuous monitoring and support.