Mining quantitative association rules in large relational tables

摘要: We introduce the problem of mining association rules in large relational tables containing both quantitative and categorical attributes. An example such an might be "10% married people between age 50 60 have at least 2 cars". deal with attributes by fine-partitioning values attribute then combining adjacent partitions as necessary. measures partial completeness which quantify information lost due to partitioning. A direct application this technique can generate too many similar rules. tackle using a "greater-than-expected-value" interest measure identify interesting output. give algorithm for Finally, we describe results approach on real-life dataset.