FOODS: Ontology-based knowledge graphs for forest observatories

Hamed, Naeima ORCID: https://orcid.org/0000-0002-2998-5056, Rana, Omer ORCID: https://orcid.org/0000-0003-3597-2646, Orozco Ter Wengel, Pablo ORCID: https://orcid.org/0000-0002-7951-4148, Goossens, Benoit ORCID: https://orcid.org/0000-0003-2360-4643 and Perera, Charith ORCID: https://orcid.org/0000-0002-0190-3346 2022. FOODS: Ontology-based knowledge graphs for forest observatories. [Technical Report].

Preview

PDF - Accepted Post-Print Version Download (34MB) | Preview

Abstract

We propose the Forest Observatory, a linked datastore, to represent knowledge from wildlife data. It is a resource that semantically integrates data silos and presents them in a unified manner. This research focuses on the forest of the Lower Kinabatangan Wildlife Sanctuary (LKWS) in Sabah, Malaysian Borneo. In this region, wildlife research activities generate a variety of Internet of Things (IoT) data. However, due to the heterogeneity and isolation of such data (i.e., data created in different formats and stored in separate locations), extracting meaningful information is deemed time-consuming and labour-intense. One possible solution would be to integrate these data using semantic web technologies. As a result, data entities are transformed into a machine-readable format and can be accessed on a single display. This study created a semantic data model to integrate heterogeneous wildlife data. Our approach developed the Forest Observatory Ontology (FOO) to lay the foundation for the Forest Observatory. FOO modelled the IoT and wildlife concepts, established their relationships, and used these features to link historical datasets. We evaluated FOO’s structure and the Forest Observatory using pitfalls scanners and task-based methods. For the latter, a use case was assigned to the Forest Observatory, querying it before and after reasoning. The results demonstrated that our Forest Observatory provides precise and prompt responses to complex questions about wildlife. We hope our research will aid bioscientists and wildlife researchers in maximising the value of their digital data. The Forest Observatory can be expanded to include new data sources, replicated in various wildlife sanctuaries, and adapted to other domains.

Item Type:	Monograph (Technical Report)
Status:	In Press
Schools:	Computer Science & Informatics Biosciences
Subjects:	Q Science > Q Science (General)
Uncontrolled Keywords:	Wildlife data, Internet of Things, Ontology, Linked data, Search and Discovery, SPARQL, Reasoning. Links: Forest Observatory Ontology: https://naeima.github.io/Forest-Observatory-Ontology/ and https://bioportal.bioontology.org/ontologies/FOO.Demo: https://youtu.be/UZYA70ph5kE.
Date of First Compliant Deposit:	11 November 2024
Last Modified:	11 Nov 2024 11:03
URI:	https://orca.cardiff.ac.uk/id/eprint/153362

Actions (repository staff only)

Edit Item