Automatic verification of the conceptual model and its documentation

Abstract

By using background knowledge of general and specific domains, and by processing a new natural language corpus, experts are able to produce a conceptual model lor some specific domain. In this paper, we present a model that tries to capture some aspects of this conceptual modeling process. This model is functionally organized into two information processing streams: one reflects the process of formal concept lattice generation from the domain conceptual model: the other reflects the process of formal concept lattice generation from the domain documentation. It is expected that similarity between those concept lattices reflects similarity between documentation and the conceptual model. In addition to this process of formal verification, a set of natural language processing artifacts are created. Those artifacts then can be used for the development of information system natural language interfaces. To demonstrate it, an experiment of concept identification from natural language queries is provided at the end of lhis paper.