Ecology and Society Ecology and Society
E&S Home > Vol. 25, Iss. 4 > Art. 45 > Abstract Open Access Publishing 
Putting machine learning to use in natural resource management—improving model performance

Ulrich J Frey, German Aerospace Center, Institute of Engineering Thermodynamics, Stuttgart, Germany

DOI: http://dx.doi.org/10.5751/ES-12124-250445

Full Text: HTML   
Download Citation


Abstract

Machine learning models have proven to be very successful in many fields of research. Yet, in natural resource management, modeling with algorithms such as gradient boosting or artificial neural networks is virtually nonexistent. The current state of research on existing applications of machine learning in the field of social-ecological systems is outlined in a systematic literature review. For this purpose, a short introduction on fundamental concepts of neural network modeling is provided. The data set used, a prototypical case study collection of social-ecological systems—the common–pool resources database from the Ostrom Workshop—is described. I answer the question of whether neural networks are suitable for the kind of data and problems in this field, and whether they or other machine learning algorithms perform better than standard statistical approaches such as regressions. The results indicate a large performance gain. In addition, I identify obstacles for adapting machine learning and provide suggestions on how to overcome them. By using a freely available data set and open source software, and by providing the full code, I hope to enable the community to add machine learning to the existing tool box of statistical methods.

Key words

comparability; gradient boosting; machine learning; natural resource management; neural networks; social-ecological systems

Copyright © 2020 by the author(s). Published here under license by The Resilience Alliance. This article is under a Creative Commons Attribution-NonCommercial 4.0 International License. You may share and adapt the work for noncommercial purposes provided the original author and source are credited, you indicate whether any changes were made, and you include a link to the license.

Top
Ecology and Society. ISSN: 1708-3087