Integrating Multi-Modal Spatial Data using Knowledge Graphs – a Case Study of Microflora Danica

Abstract

Integrating semantically related, multi-modal, heterogeneous data sources is challenging, especially if one of the modalities includes spatial data, such as field measurements organized in geographical grids. Since geographical grids can have different rotations, be translated along one or more axes, or have different resolutions, a particular challenge when integrating such data is to reduce the information loss from projecting different grids into a common format. In this paper, we study this problem and sketch a method for integrating such spatial data using knowledge graphs. We discuss this solution in the context of a real-world use case, where we integrate geographically annotated microbial data (Microflora Danica) as well as environmental data to enable joint analysis. The first results of our experiments show that our method reduces the information loss compared to baseline methods.