Mining Frequent Item sets on Large Scale Temporal Data

Frequent pattern mining has become an important data mining technique that is mainly focused in many research fields. Frequent patterns are the patterns that appear frequently in the dataset. Several algorithms have been proposed to mine all the frequent item sets in the dataset. These algorithms differ from other algorithms by reducing the number of items in the dataset and in the generation of candidate sets. This paper attempts to propose a new data-mining algorithm for mining all the frequent item sets based on the temporal data which contains the time-stamping information. We propose an efficient algorithm for mining frequent item sets by extending FP-Growth algorithm based on temporal data. Here the concept is that by avoiding the candidate generation. Only the sub-databases are tested.