Ontology Based Multi Agent System For Human Disease Studies

This thesis makes three unique contributions, namely: (1) It developed a ten-phase Onto-Agent design methodology which is the first attempt in the world-wide research community to unify both ontology modelling and a multi-agent technique in a formal design framework. The Onto-Agent methodology comprises two interconnected processes: an ontology design process and a multi-agent system design process. The multi-agents design is based on the ontology which is used for disease query formulation, disease information retrieval, and disease information integration. The use of this ontology facilitates: information collection from multiple information resources, data sharing, and management of scientific knowledge about disease, in a timely manner. (2) The thesis developed an ontological model that represents a general knowledge of human diseases and contains four subontologies: Disease Types, Disease Symptoms, Disease Causes and Disease Treatments. This proposal of a Human Disease Ontology on such a large scale is the first of its kind in the world-wide field of study of diseases. (3) The thesis provided a theoretical base to prove the ontology design through the use of a Set Theory, and a definition of the dynamics of Ontology-based Multi-agent systems through the use of a Coalgebraic approach and Coinduction principle. This is the first time in the literature that system dynamics is approached in this way. The methodology developed from this thesis will be suitable not only for the study of human diseases, but can also bc adapted to any other knowledge domains.