Discovery of approximate dependencies from proximity-based fuzzy databases

We present a data mining technique for discovering approximate dependencies from proximity-relation-based fuzzy databases. A proximity-relation-based fuzzy data model was proposed by Shenoi (1989) as an extension to the similarity-relation-based fuzzy data model. It is most suitable for describing analogical data over discrete domains, in addition to fuzzy-set-based fuzzy data models. Approximate dependency is an extension of functional dependency such that equality of tuples is extended and replaced with the notion of equivalence class. Many extensions of functional dependencies have been discussed in the level of conceptual viewpoints. We adopt a level-wise mining technique originated by Huhtala (1998) for the search of all possible nontrivial minimal approximate dependencies. The approximate dependency we define can capture more real-world integrity constraints on fuzzy databases