Affiliations: College of Information Science & Engineering, Northeastern University, Shenyang, China | School of Software, Northeastern University, Shenyang, China
Note: [] Corresponding author. Z.M. Ma, PhD, Professor, IEEE Senior Member, College of Information Science and Engineering, Northeastern University, Shenyang 110819, China. E-mail: mazongmin@ise.neu.edu.cn
Abstract: Description logics have been studied in depth in data modeling. This paper proposes a description logic approach for representing and reasoning on fuzzy ER models. Firstly, an analysis about how to choose suitable description logic as a modeling language for fuzzy ER models is given, which motivates the use of the fuzzy description logic called FDLR (the fuzzy extension of DLR) for the purpose of this paper. Then, a formal definition of fuzzy ER models is proposed. Furthermore, based on the FDLR and the formalization of fuzzy ER models, we propose a complete formal approach for representing fuzzy ER models with FDLR, i.e., transforming fuzzy ER models into FDLR knowledge bases at both terminological (schema) and assertional (instance) levels. On this basis, following the proposed approach, we implemented a prototype tool FER2FDLR. Finally, based on the transformed FDLR knowledge bases, we investigate how to reason on fuzzy ER models with FDLR, and the reasoning tasks of fuzzy ER models can be reasoned through the reasoning mechanism of FDLR.
Keywords: Fuzzy ER model, fuzzy description logic, representation, reasoning
