Knowledge elicitation and representation in a normative approach-a case study in diagnosis of plant diseases in Egypt

Knowledge becomes a pivotal factor for intelligence to support decision making. Elicitation, representation and use of knowledge are major areas of research in the field of Artificial Intelligence, leading to development of knowledge bases and expert systems. Knowledge base system (KBS) development is complex, expensive, and time consuming process that needs to be applied in an organized manner. There are powerful tools for eliciting and representing KBSs. Using these tools becomes a very difficult task for users without specific knowledge in Artificial Intelligence. The aim of our research is to use a normative approach to elicit and represent the knowledge. To achieve this we build an elicitation and representation tool NormEST which allows the domain experts and the non-technical users are able to build their own KBSs. NormEST can be considering as a new approach in eliciting and representing the knowledge. NormEST is designed based on the concepts of Organisational Semiotics and evaluated by group of agriculture domain experts, the Delphi technique used as evaluation method to illustrate the top 10 ranked benefits of using NormEST. NormEST is demonstrated using a case study in fungal diseases in cucumber in Egypt. The results indicate that the NormEST is able to facilitate effective elicitation of parameters needed for an agriculture diagnostic experts system.