Frequent Itemset Mining using Parallel Processing on Temporal Data

COMPUTER ENGINEERING (DATA MINING)

Frequent pattern extraction is important technique in data mining domain. Lot of work has been done on pattern mining in literature. The work focuses on pattern extraction on static dataset. The whole dataset is used and frequent itemset are extracted. Temporal data is the data containing timestamp information. Timestamp based pattern extraction affects the results. The proposed work focuses on frequent itemset extraction on temporal data. The whole data is partitioned in time cubes based on the timestamp information. For itemset extraction apriori algorithm is applied on each time cube. A density threshold is used to overcome the overestimation problem of time period. The technique enlists the patterns those are occurred in particular time interval. The work improves the system efficiency with parallel thread working. Comparative results are taken on the basis on time and memory.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.