Reifying Model Integration Abilities from Natural Language

The paper is a study on natural language (NL) from the viewpoint of the conceptual model integration. The universality, unifying abilities, natural extensibility, logic and reusability of NL, as well as its morphological and syntactic stability, made the modelers think of transposing these features to object and process modeling. A correspondence is established here between the main concepts and relationships in NL and the basic concepts and relationships which the designers use in different models like: object, process, dataflow and workflow models. Two main directions are proposed for the study on the integration abilities of NL: (1) the morphological and syntactic categories and (2) the semantic and intersentential relations between objects and actions. These categories and relations are supposed to support the seamless integration of the models, as the background either of a meta-model with linguistic features or of a unifying translation algorithm from NL to several conceptual models.