Ontology Based Agriculture Data Mining using IWO and RNN

Abstract

An ontology is a machine-interpretable formal description of domain knowledge. In current years, ontologies have risen to prominence as a key tool for demonstrating domain knowledge and a key element of several knowledge management systems, decision-support systems (DSS) and other intelligent systems including in agriculture. However, a study of the current literature on agricultural ontologies suggests that the majority of research that suggest agricultural ontologies lack a clear assessment mechanism. This is unwanted because this is impossible to assess the value of ontologies in research and practise without well-structured assessment mechanisms. Furthermore, relying on such ontologies and sharing them on the Semantic Web or amongst semantic-aware apps is problematic. This paper presents a framework for selecting appropriate assessment techniques for Ontology Based Agriculture Data Mining utilizing Invasive Weed Optimization (IWO) and Re-current Neural Network (RNN) that appears to be absent from most recent agricultural ontology research. The framework facilitates the selection of relevant evaluation techniques for a particular ontology based on its intended user.