TreeNetGaps
Gestion de projet
Suppléance
Collaborateurs du projet
Project Description ¶
| In a recent WSL research project (deepT - internal grant no. 202011N2099) we developed a machine learning model for gap-filling multi-channel time series data. We would now like to add a module or toolbox that is based on this model to the existing automated near real-time TreeNet data acquisition infrastructure. The new tool would provide an additional option to the users of TreeNet dendrometer data to automatically fill the gaps in the time series using artificial intelligence. The existing model first must be improved using newly available data. It then has to be programmed in R, the native language used in the TreeNet software. Thereafter, the code must be inserted into the existing pipeline and offered as an option to the end-users of the data. This will also involve adapting the data to the input and output requirements of the model. |
References ¶
Mirko Lukovic, Roman Zweifel et al. Journal of the Royal Society Interface, 19, 20220349 (2022).