Ganitha Piyasa: Effective Lesson Delivery Method for Graphical Dyscalculia Students

Many children face learning disabilities, each requiring unique solutions such as physical exercises, teaching techniques, and behavioral interventions. A specific challenge lies in Graphical Dyscalculia, which hinders children's mathematical abilities. This study presents the “Ganitha Piyasa” mobile application, tailored to help children with Graphical Dyscalculia enhance their numerical proficiency and number writing skills. The application provides a structured learning environment with fifteen progressively challenging levels. It offers performance-based feedback and evaluation, alongside comprehensive data collection recording task-specific details, completion times, and accuracy. An Artificial Neural Network (ANN) model, initially trained with the MNIST handwritten digits data set in Keras, accurately assesses number writing. Converted into a Tensor Flow Lite file, this model ensures efficient real-time evaluation. A Random Forest Classifier, utilizing metrics like average writing time and success probabilities, predicts a child's improvement in number writing skills. This aids personalized learning path-ways. The study involved 50 children who were diagnosed with Dyscalculia aged seven to ten, demonstrating the application's effectiveness. “Ganitha Piyasa” boasts an intuitive interface that encourages engagement and parental involvement. Parents can monitor their child's progress through earned achievement stars, motivating sustained usage. The application's methodology aligns with medical expertise, ensuring effective enhancement of children's numerical skills.