Study of Mining Frequent Patterns at Various Levels of Abstraction

Abstract

The discovery of interesting association relationships among huge amounts of business transaction records can help in many business decision making process, association rules is one of the main popular pattern discovery techniques in data mining (KDD).The problem of dis-covering association rules has received considerable research attention and several algorithms for mining frequent pattern at primitive and multi-ple level have been developed. In this paper, we have studied various association rule mining algorithms like primitive association rule mining, generalized association rule mining and multilevel association rule mining. Mining primitive association rules helps in finding general knowl-edge considers all items at single level. Generalized association rule mining provides extra knowledge as sibling associations and even cross-parent associations. Multilevel association rule mining algorithm takes care of analyzing different level wise knowledge.

Keywords: Primitive association rules, Multiple level association rules, Generalized association rules, Data mining, Support, Confidence.