Publication
Machine Learning on Simulated and Real Farm Data based on an Ontology-Controlled Data Infrastructure
Alexander Muenzberg; Christian Troost; Daniel Martini; Francisco Mendoza; Rajiv Srivastava; Thomas Berger; Liv Seuring; Nils Reinosch; Thilo Streck; Ansgar Bernardi
In: K. Hinkelmann; A. Martin (Hrsg.). Proceedings of the AAAI 2022 Spring Symposium on Machine Learning and Knowledge Engineering for Hybrid Intelligence. AAAI Spring Symposium on Machine Learning and Knowledge Engineering for Hybrid Intelligence (AAAI-MAKE-2022), March 21-23, Palo Alto, California, USA, AAAI Spring Symposium Series, AAAI, 2022.
Abstract
The SimLearn project uses semantic information processing and machine learning to combine processmodel simulation data with real-world farm data in order to to support decision making in agriculture. This paper describes a system architecture for high-performance extraction, storage, processing and machine learning of data extracted from simulation models (Big Data) or collected from farmers. Ontologies are used to link data to expert knowledge. Neural networks are pre-trained on simulation output to be later employed in prediction of decision-relevant outcomes (e.g. crop growth, soil mineralization) after transfer learning from observed farm data.