Functional dependency with degree of satisfaction (FDd) is an extended notion in data modeling, and reflects a type of integrity constraints and business rules on attributes, mainly for massive databases, in which incomplete data such as noise, null and imprecision may exist. While existing approaches are considered effective in general, attempts for further improvement in efficiency are deemed meaningful and desirable as far as knowledge discovery is concerned. This paper focuses on discovering (FDd)s as a form of useful semantic knowledge, aiming at providing an enhancement to the FDd mining process in a more efficient manner. In doing so, properties of FDd are in-depth investigated along with a measure for degree of distinctness. Subsequently, a number of optimization strategies are developed for pre-processing, which are then incorporated into the mining process, giving rise to an enhanced approach for mining functional dependency with degree of satisfaction, namely e-MFDD. Finally, data experiments revealed that e-MFDD significantly outperformed the original approach without pre-processing.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.