Downloads: 1 | Views: 9 | Weekly Hits: ⮙1 | Monthly Hits: ⮙1
Research Paper | Computer Science and Information Technology | Saudi Arabia | Volume 13 Issue 7, July 2024 | Rating: 3.6 / 10
An Ontology Based Approach for Plant Pests and Diseases Diagnosing and Control
Amani F. Alharbi | Muhammad A. Aslam | Khalid A. Asiry | Naif R. Aljohani | Souad A. Baowidan
Abstract: In recent years, ontologies have gained popularity in the domain of agriculture as techniques for semantic knowledge management and modeling. Their ability to allow data integration and enable semantic interoperability across heterogeneous sources is a major advantage in smart agriculture. However, existing ontologies? computational power over semantic expressivity and inference of hidden knowledge has gaps that need to be bridged. Specifically, for these ontologies to be useful in supporting agriculture decisions, all related factors should be considered; moreover, they should be more computationally rigorous to detect the complex relationships hiding between these factors. To address these issues, we present a Plant Diseases and Pests Ontology (PDP-O). This ontology captures all related concepts, terms, and semantic relations about plant diseases and pests, and defines a schema to describe these elements on the basis of formal semantics. To enhance PDP-O?s reasoning capabilities we adopted three strategies: (1) describing PDP-O classes using logical description; (2) modeling to enrich the meaning of PDP-O relationships; and (3) adding SWRL rules to represent complex relationships that cannot be defined through OWL DL. As proof of concept, we provide a brief use case to identify a disease affecting date palm crops. To demonstrate PDP-O?s efficacy and validity, automatic consistency checking was performed, and the results obtained show that PDP-O is both coherent and consistent. This ontology can benefit both domain stakeholders and systems developers. First, it contains information on various factors relating to plant diseases and pests (such as symptoms, most susceptible varieties, outbreak time, preferable environmental factors, and treatment or control measures), empowering agricultural stakeholders to improve farm management and enhance their adherence to best practice. Second, it provides an in-depth computationally enriched schema, based on which developers can develop more intelligent decision-support systems and simplify their integration of data from heterogeneous sources.
Keywords: Ontology, RDF/OWL, Machine reasoning, Agriculture, Pest recognition, Pest control
Edition: Volume 13 Issue 7, July 2024,
Pages: 243 - 253