Granule Oriented Data Warehouse Model

摘要

It is a big challenging issue to assure the quality of multidimensional association rules due to the complexity of the association between attributes. Granule mining divides data attributes into multi-tiers and compresses them into granules based on these tiers. Useful rules then can be justified according to the relationship between granules in tiers. Meanwhile, data warehousing is an ideal platform in handling enormous data that helps data mining to focus on representations of rules that best fit users' interests. In this paper, a granule oriented data warehouse model is proposed where the association mappings are implemented to represent the relationship between granules in multi-tiers. Experiments show that the proposed solution achieves encouraging performance.