This is a small time-series challenge I did during my master's. All credit for the challange idea goes to Dr. Sharon Ong, Department of Cognitive Science and Artificial Inteligence, Tilburg University.
Update: My implementation of Rocket managed to win 2nd place in the competition :)
The objective of this challange is to classify whether a child's handwriting is affected by dysgraphia. The data comes from the study by Drotár and Dobeš, 2020 and is avaliable here. It was collected using a using a WACOM Intuos Pro Large tablet. The features are numeric and represent the below over time:
- pen movement in the x-direction,
- pen movement in the y-direction
- whether the pen was on the surface (1) or in the air (0)
- the pressure of the pen on the tablet surface
- the azimuth of the pen on the tablet surface
References Drotár, P., Dobeš, M. Dysgraphia detection through machine learning. Sci Rep 10, 21541 (2020). https://doi.org/10.1038/s41598-020-78611-9