ITFP: Incremental TFP for mining frequent patterns from large data sets

Previous studies indicate that FP-Growth has a fast performance while Apriori-TFP is more efficient in terms of used memory. Based on those characteristics, in this paper, we proposed an Apriori-TFP based incremental frequent pattern mining algorithm that can search efficiently within limitation of memory and the further classification work base on those patterns. Especially, the concept of pre-infrequent patterns pruning and use of two different minimum supports, it made the algorithm be possible to handle the problem of mining frequent patterns from incrementally increased, large size of data sets.