Technical note:  Incorporating expert domain knowledge into causal structure discovery workflows

Mäkelä, Jarmo; Melkas, Laila; Mammarella, Ivan; Nieminen, Tuomo; Chandramouli, Suyog; Savvides, Rafael; Puolamäki, Kai

doi:https://doi.org/10.5194/bg-19-2095-2022

Articles | Volume 19, issue 8

https://doi.org/10.5194/bg-19-2095-2022

© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/bg-19-2095-2022

© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 19, issue 8

Technical note

|

19 Apr 2022

Technical note |

| 19 Apr 2022

Technical note: Incorporating expert domain knowledge into causal structure discovery workflows

Jarmo Mäkelä, Laila Melkas, Ivan Mammarella, Tuomo Nieminen, Suyog Chandramouli, Rafael Savvides, and Kai Puolamäki

Data sets

Interactive Causal Structure Discovery with Hyytiälä measurements L. Melkas, J. Mäkelä, S. Chandramouli, I. Mammarella, T. Nieminen, K. Puolamäki, and R. Savvides https://doi.org/10.5281/zenodo.6451931

Model code and software

Interactive Causal Structure Discovery with Hyytiälä measurements L. Melkas, J. Mäkelä, S. Chandramouli, I. Mammarella, T. Nieminen, K. Puolamäki, and R. Savvides https://doi.org/10.5281/zenodo.6451931

Short summary

Causal structure discovery algorithms have been making headway into Earth system sciences, and they can be used to increase our understanding on biosphere–atmosphere interactions. In this paper we present a procedure on how to utilize prior knowledge of the domain experts together with these algorithms in order to find more robust causal structure models. We also demonstrate how to avoid pitfalls such as over-fitting and concept drift during this process.