Interestingness-Based Interval Merger for Numeric Association Rules

We present an algorithm for mining association rules from relational tables containing numeric and categorical attributes. The approach is to merge adjacent intervals of numeric values, in a bottom-up manner, on the basis of maximizing the interestingness of a set of association rules. A modi cation of the B-tree is adopted for performing this task e ciently. The algorithm takes O(kN ) I/O time, where k is the number of attributes and N is the number of rows in the table. We evaluate the e ectiveness of producing good intervals.

• Type: Conference
• Year: 1998
• Where: KDD
• Authors: Ke Wang, Soon Hock William Tay, Bing Liu
• Keywords: Adjacent Intervals, Association Rules, Categorical Attributes, Data Mining, KDD 1998