What is Generator Item-Set

Association Rule Mining (ARM) is one of the important data mining tasks that has been extensively researched by data-mining community and has found wide applications in industry. An Association Rule is a pattern that implies co-occurrence of events or items in a database. Knowledge of such relationships in a database can be employed in strategic decision making in both commercial and scientific domains. A typical application of ARM is market basket analysis where associations between the different items are discovered to analyze the customer’s buying habits. The discovery of such associations can help to develop better marketing strategies. ARM has been extensively used in other applications like spatial-temporal, health care, bioinformatics, web data etc (Hipp J., Güntzer U., Nakhaeizadeh G. 2000). An association rule is an implication of the form X ? Y where X and Y are independent sets of attributes/ items. An association rule indicates that if a set of items X occurs in a transaction record then the set of items Y also occurs in the same record. X is called the antecedent of the rule and Y is called the consequent of the rule. Processing massive datasets for discovering co-occurring items and generating interesting rules in reasonable time is the objective of all ARM algorithms. The task of discovering co-occurring sets of items cannot be easily accomplished using SQL, as a little reflection will reveal. Use of ‘Count’ aggregate query requires the condition to be specified in the where clause, which finds the frequency of only one set of items at a time. In order to find out all sets of co-occurring items in a database with n items, the number of queries that need to be written is exponential in n. This is the prime motivation for designing algorithms for efficient discovery of co-occurring sets of items, which are required to find the association rules. In this article we focus on the algorithms for association rule mining (ARM) and the scalability issues in ARM. We assume familiarity of the reader with the motivation and applications of association rule mining