Fuzzy statistical dependency and normalisation in fuzzy statistical database

摘要

Most of the statistical databases developed so far use precise data. But in the real world, most of the data is imprecise in nature. Fuzzy set theory has proven to be a good choice for accomplishing the task of imperfect data particularly in statistical database and fuzzy statistical database has arisen as a result. In this paper, we introduce the concept of fuzzy statistical dependency, a dependency between the attributes of fuzzy statistical database which allows the database developer to develop a consistent logical modelling tool for fuzzy statistical database. We also discuss the fuzzy statistical equality for comparison of attributes in fuzzy statistical database. To eliminate the redundancy in fuzzy statistical database, fuzzy statistical normal forms are defined which present a framework for the database developer to maintain consistency in the fuzzy statistical database. Algorithms for implementation of fuzzy statistical normal forms are also discussed.