Extending Fuzzy Relational Database With Distance Relation And The Clustering Algorithms
Yiyong Xiao,Ikou Kaku,Wenbing Chang
2008-01-01
Abstract:It is generally considered that the fuzzy database is the next generation of the conventional database because the information in the real world is always vague or ambiguous. The theory of fuzzy relational algebraic operations, however, is not so solid as that of conventional relational data model, which has been considered to be the main reason of preventing this theory from widely being applied. For two fuzzy values, despite of the fuzzy-set-based or the possibility-based type, there does not exist a common way to calculate their closeness or distance measure, even though the closeness relations between single elements on scalar domain have been pre-defined in advance. In this paper, we first present a framework of incorporating the possibility-based fuzzy information in a conventional relational database structure, as definition 6 and 7, which uniforms the expressing format of determinate values and fuzzy values, and provides a way to handle the determinate value as a special fuzzy value. And then, we extend the often used definition of closeness relation on fuzzy database to distance relation, based on which the most important contribution of this paper, the distance function of fuzzy values, has been presented and proofed, as theorem 1 and Theorem 2. Numeric example has been given to validate the validity. Clustering algorithm on data objects in fuzzy database is therefore promoted.