Flexible evaluation of textual labels in conceptual models

Abstract

This paper introduces a flexible and generic approach to define customised style rules for labels in conceptual models. A rule-based language is presented which can express style rules in a flexible and context-specific way. The formalised style rules are used to analyse and evaluate textual labels of model elements. To analyse a textual label, a combination of standardised natural language processing tools such as a part-of-speech tagger or a named entity recogniser are used. With the help of these techniques, custom-defined information entities can be extracted from the model.