Name
Abstract | ||
|---|---|---|
Spatial interpolation is the task to predict a measurement for any location in a given geographical region. To train a prediction model, we assume to have point-wise measurements for various locations in the region. In addition, it is often beneficial to consider historic measurements for these locations when training an interpolation model. Typical use cases are the interpolation of weather, pollution or traffic information. In this paper, we introduce a new type of model with strong relational inductive bias based on Message Passing Networks. In addition, we extend our new model to take geomorphological characteristics into account to improve the prediciton quality. We provide an extensive evaluation based on a large real-world weather dataset and compare our new approach with classical statistical interpolation techniques and Neural Networks without inductive bias. |
Year | DOI | Venue |
|---|---|---|
2019 | 10.1109/ICDMW.2019.00030 | 2019 International Conference on Data Mining Workshops (ICDMW) |
Keywords | Field | DocType |
graph neural networks,message passing networks,spatial interpolation | Data mining,Inductive bias,Use case,Multivariate interpolation,Computer science,Interpolation,Graph neural networks,Artificial neural network,Message passing | Conference |
ISSN | ISBN | Citations |
2375-9232 | 978-1-7281-4897-7 | 0 |
PageRank | References | Authors |
0.34 | 4 | 8 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Evgeniy Faerman | 1 | 0 | 0.68 |
| Manuell Rogalla | 2 | 0 | 0.34 |
| Niklas Strauß | 3 | 0 | 0.34 |
| Adrian Krüger | 4 | 0 | 0.34 |
| Benedict Blümel | 5 | 0 | 0.34 |
| Max Berrendorf | 6 | 0 | 0.68 |
| Michael Fromm | 7 | 0 | 0.68 |
| Matthias Schubert | 8 | 740 | 55.53 |