Publication Date

2006

Degree Type

Master's Project

Degree Name

Master of Science (MS)

Department

Computer Science

Abstract

Bitmap is an extremely efficient way of representing data, but the drawback is that the order of data is fixed in a bitmap. Granular computing is a new theory that frees the bitmap method from fixed order of data in the same manner as linear algebra frees the matrix theory from a fixed basis. To obtain meaningful information using data mining techniques has been a central idea in recent database applications [2]. One of the core techniques in data mining is to find associations (undirected association rules) between attribute values [4]. The complexity of finding associations is often very high. In this work, a data warehouse is constructed utilizing bitmap fundamentals. Classic method of obtaining association rules is implemented along with the granular computing method of obtaining association rules. The performances from classic method and granular computing method is carefully analyzed and compared.

Download

Share

COinS