Fast Mining Of Association Rules To Identify the Audience for Ad

Given the importance of large databases in today's world, the problem of discovery of useful knowledge and association rules isparticularly important in order to understand the audience .Most existing studies on association rules discovery focused onfinding the association rules between all items in a large database that satisfy user-specified minimum confidence and support.In practice, in order to identify the audience in the database, we are often interested in finding association rules involving onlysome specified items. Meanwhile, based on the search results in former queries, users tend to change the minimal confidenceand support requirements to obtain suitable number of rules. Under these constraints, the existing mining algorithms can notperform efficiently due to high and repeated disk access overhead. In this research, in order to understand the audience with theaim of directing the advertisement , we present a novel mining algorithm named creating sub vector(CS) Which can efficientlydiscover the association rules. At most one scan of the database is needed for each query; hence, the disk access overhead canbe reduced substantially and the query be responded quickly.We have performed extensive expriments and compared theperformance of the algorithm with one of the best existing algorithms.

Keywords:

Data mining , Association rules , Subvector , Cluster