Generic fuzzy reasoning nets as a basis for reverse engineering relational database applications

Abstract

Object-oriented technology has become mature enough to satisfy many new requirements coming from areas like computer-aided design (CAD), computer-integrated manufacturing (CIM), or software engineering (SE). However, a competitive information management infrastructure often demands to merge data from CAD-, CIM-, or SE-systems with business data stored in a relational system. One approach for seamless integration of objectoriented and relational systems is to migrate from a relational to an objectoriented system. The first step in this migration process is reverse engineering of the legacy database. In this paper we propose a new graphical and executable language called Generic Fuzzy Reasoning Nets for modelling and applying reverse engineering knowledge. In particular, this language enables to define and analyse fuzzy knowledge which is usually all what is available when an existing database schema has to be reverse engineered into an object-oriented one. The analysis process is based on executing a fuzzy petri net which is parameterized with the fuzzy knowledge about a concrete database application.