A framework for the semantic interoperability of datasets: a case study in agriculture

• Grant number: 22/08385-8
• Support Opportunities: Scholarships abroad - Research Internship - Doctorate
• Start date: January 10, 2023
• End date: January 09, 2024
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computer Systems
• Principal Investigator: Antonio Mauro Saraiva
• Grantee: Filipi Miranda Soares
• Supervisor: Luis Ferreira Pires
• Host Institution: Escola Politécnica (EP). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Institution abroad: University of Twente (UT), Netherlands
• Associated to the scholarship: 21/15125-0 - A metadata model for interoperability of agricultural databases, BP.DR

Abstract

One of the main characteristics of the big data paradigm is the unstructured way in which data are published. Big data applications have great potential to be used in digital agriculture, but one may find it difficult to work with unstructured data since most datasets do not provide an accurate description of the data semantics, hampering interpretation and reuse. Consequently, semantic interoperability between datasets may be jeopardized, since the datasets do not share the same metadata concepts in their records. To address that problem, this research project proposes the development of metadata models for the semantic interoperability of datasets in the agricultural domain, focusing on the subdomain of agricultural commodities. To achieve our research objective, the Design Science research principles are adopted as the methodological approach, since it allows the development of research artifacts for the purpose of acquiring knowledge. Two main artifacts are going to be created: 1) a metadata schema for the domain; 2) a metadata semantic mapping among different schemas. The first artifact aims at the publication of agricultural commodities data as semantically structured data, and the second artifact aims at enabling interoperability among datasets by establishing a semantic equivalence between metadata terms used in each dataset. In the last stage of the research, the general applicability of the proposed metadata models will be challenged by applying them in another related field, namely food and nutrition. Finally, the metadata schema will be validated by domain specialists. (AU)