Periodic knowledge discovery through parallel paradigm

Temporal association rules are largely different from traditional association rules by the fact that temporal association rules attempt to model temporal relationships in the data. Effective gain in any business is possible to achieve due to the adaptive knowledge which demands customized rules for specific conditions. Several parallel algorithms are useful to extract frequent patterns from large databases. This paper proposes a novel methodology for extracting calendric association rules and hence the general rules for a timestamp transactional database through modified Parallel Compact Pattern Tree construction strategy. The same has been demonstrated through mushroom dataset and synthetic temporal transactions.