ACIS 2024 Proceedings
Abstract
Ontology-based information systems are crucial for providing farmers with actionable, context-specific information. However, the rapidly evolving nature of agricultural information made the ontology information inaccurate and incomplete, limiting its usefulness to farmers. On the other hand, manually updating the ontology with evolving information is inefficient and labour-intensive. To address these real-world problems and manage their complexity, the authors adopted the Design Science Research (DSR) methodology. As a result of following the DSR cycles, the study developed a semi-automated system. The system has automated the ontology learning process, in which ontology entities are extracted from published documents, by employing the Large Language Model (LLM) technologies. However, expert validation and ontology curation remain manual to maintain the high accuracy of the ontology. The functional and utility tests demonstrate the feasibility and effectiveness of the developed solution. The ongoing improvements guided by DSR are being continued to ensure its relevance in real-world scenarios.
Recommended Citation
Wilson, Shyama; Ginige, Athula; and Goonatilake, Jeevani, "Design Science Research Approach for Ontology Development in Agriculture: Utilising advances of LLM for Automated Entity Extraction" (2024). ACIS 2024 Proceedings. 150.
https://aisel.aisnet.org/acis2024/150