Ontology Based Model Integration
Xianming Liu · Dianzi xuebao · 2009
Model plays very important role in Decision Support System(DSS).How to integrate models and improve models collaboration have always been the key research spot in DSS field.Model parameters′ representation and matching are the core problems in model integration Due to the difficulties in describing the semantic of the input,output parameters,the intelligence,extensibility,openness and understanding of automatic model integration are not satisfied and need to be improved.An ontology-based program-logic-mix(PLM)model representation is proposed,which is easily-implemented,reasoning-supported,has high openness and extensibility.Ontology is used to describe the semantics of parameters.Based on this semantic match,model integration strategies are designed to create understandable decision problem solutions.Formal definition of model integration is given and relevant theorems are proved.