Incremental Association Rule Mining With a Fast Incremental Updating Frequent Pattern Growth Algorithm

One of the most challenging tasks in association rule mining is that when a new incremental database added to an original database, some existing frequent itemsets may become infrequent and vice versa. As result, previous rules invalid emerge. We designed new, more efficient approach for using Fast Incremental Updating Frequent Pattern growth algorithm (FIUFP-Growth), Conditional tree (ICP-tree), compact sub-tree suitable itemsets. This retrieves have already been mined from their support counts then use them efficiently mine updated ICP-tree, reducing number rescans database. Our reduced usages resource time unnecessary construction compared individual FP- Growth, FUFP-tree maintenance, Pre-FUFP, FCFPIM algorithms. From results, at 3% minimum threshold, average execution pattern our performs 46% faster than FUFP-tree, FCFPIM. experimental findings directly benefit designers developers computer business intelligence methods.