Dysgraphia is a learning disorder characterized by difficulties with writing, particularly with spelling, grammar, punctuation, and handwriting. Handwriting analysis plays a crucial role in detecting dysgraphia. It is imperative to detect dysgraphia at early stages to improve individuals writing skills, promote better self-esteem and mental well-being, and develop a technology-based solution to help individuals with dysgraphia overcome the challenges that they struggle with organizing their thoughts or become anxious. The aim is to develop a dysgraphia detection system using supervised machine learning algorithms that build upon previous research and leverage various techniques to enhance the detection of dysgraphia. The supervised machine learning algorithms are trained and tested with various split ratios (60–40, 70–30, 80–20) on a dataset collected from children between the ages of 7 to 12 years. To ensure a robust and balanced dataset, a variety of data augmentation strategies are used to get over the limited availability of handwriting data. The developed system achieves the highest accuracy 97.61% for AdaBoost Classifier to detect dysgraphia in children's handwriting data.