Abstract
The article presents the development of an intelligent writing analysis system based on touch trajectories and spline filtering methods, focused on the diagnosis and development of graphomotor skills in preschoolers in digital writing learning. The relevance of the study is due to the need for early detection of difficulties in the formation of handwriting and motor coordination using objective digital tools, which is especially important when preparing children for school. Polygonal approximation, piecewise linear functions, and Catmull-Rom spline interpolation are used as the basis for mathematical analysis of trajectories. An algorithm for processing user input is proposed, which includes the stages of normalization, smoothing, comparison with a digital standard, and error categorization. The system allows you to take into account the specifics of the tool used (finger or stylus), distinguish between types of tasks (lines, continuous and discontinuous letters), as well as analyze spatial and temporal parameters of execution. As part of the testing, an experimental test of the effectiveness of the Dexterous Fingers application was conducted on a sample of 110 children aged 4-6 years. The results demonstrated a statistically significant improvement in graphomotor skills in the experimental group compared with the control group using paper records (p < 0.001). There was an increase in the proportion of children with a high level of skills, a decrease in the error rate and an increase in motivation to learn writing. The proposed system provides teachers with a tool for automated handwriting analysis, building individual development trajectories and selecting corrective tasks, and also creates the basis for further integration of artificial intelligence technologies into the processes of digital pedagogy and monitoring of children's development.