Abstract
A concept of a progressive system coordinates data and ideas in various leveled or fractional requests. It assists express connections among knowledge and data in a dataset in summarizing, advanced terms and assumes a significant part in the knowledge discovery with processing. Concept hierarchies are given by knowledge engineers, domain specialists, and clients, or integrated into certain data relationships. However, it very well might be alluring to naturally create a few calculated progressive systems or to adjust a few given progressive systems to explicit learning tasks. In this research paper, we consider the dynamic age difficulties and the expansion of idea hierarchies. Due to data distributions, this study, therefore, generates concept progressive systems for mathematical properties and dynamically refines known or original ideas based on learning requirements, associated datasets, and database statistics. These methods were evaluated on sizable relational databases and incorporated into the Database Learn Knowledge Discovery System. The methods for knowledge discovery in massive datasets are efficient and effective, according to experimental results.
Keyword
Discovery methods, Count propagation, Knowledge discovery in huge datasets, Attribute generalization, Concept hierarchies, Novel refinement of concept hierarchies
PDF Download (click here)
|