Conceptual models similarity computation based on cognitive science

In the process of information system conceptual modeling, it is often necessary to merge or reuse the existing concept models, to build a complex model to describe the physical world model more accurately. Conceptual models similarity computation is an important base of merging, mapping and reusing conceptual models. OntoUML based conceptual model constructed ensures the validity of conceptual model ontologically, and is easily used to measure the similarity between concepts. On the basis of the study of OntoUML modeling primitives, this paper introduced a method of system similarity computation in the cognitive science, and transformed conceptual model similarity computation into a synthetic computation of the similarity of relationship, binding, entity classes and roles. Experimental results shew that this computing method takes into account various relevant factors, the results are more reliable.