Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2503.15513 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Table of Contents:
- Dysgraphia is a key cognitive disorder impacting writing skills. Current tests often identify dysgraphia after writing issues emerge. This paper presents a set of computer games and uses machine learning to analyze the results, predicting if a child is at risk. The games focus on cognitive differences like visual attention between dysgraphic and typical children. The machine learning model forecasts dysgraphia by observing how kids interact with these games. We also create an algorithm to detect unsuitable testing conditions, acting as a preprocess to avoid mislabeling them as dysgraphia. We developed a machine learning model capable of predicting dysgraphia with 93.24% accuracy in a test group of 74 participants.