Finding the boundaries of attributes domains of quantitative association rules using abstraction: a dynamic approach
Ranjeet Kumar, Preetham Kumar, V. S. Ananthanarayana · 2007
The Need to extract the association rules from the pool of varied data, having gained increased momentum in the field of data mining, necessitates the discovery of methods to process multi-dimensional data, and find the qualitative or quantitative association rules from it by considering all the relevant fields in an efficient manner In this paper we propose an efficient and novel algorithm for finding the boundaries of attributes domains dynamically. It first builds an abstraction, called Multi-Variate Tree, in single scan of the database. During this construction the boundaries of domains of (quantitative) attributes are identified dynamically. These identified attributes with boundary values which are frequent are then used for finding association rules.