Hitting the memory wall: implications of the obvious
ACM SIGARCH Computer Architecture News
Mining frequent patterns without candidate generation
SIGMOD '00 Proceedings of the 2000 ACM SIGMOD international conference on Management of data
Sequential PAttern mining using a bitmap representation
Proceedings of the eighth ACM SIGKDD international conference on Knowledge discovery and data mining
Mining Sequential Patterns by Pattern-Growth: The PrefixSpan Approach
IEEE Transactions on Knowledge and Data Engineering
Cache-conscious frequent pattern mining on a modern processor
VLDB '05 Proceedings of the 31st international conference on Very large data bases
PAID: Mining Sequential Patterns by Passed Item Deduction in Large Databases
IDEAS '06 Proceedings of the 10th International Database Engineering and Applications Symposium
Hi-index | 0.00 |
In this paper, we propose the CCDR-PAID algorithm which is a technique to improve CPU cache utilization of sequential pattern mining more efficiently as an extension of existing CC-PAID which is our previous work. Compared to PAID, CC-PAID improves temporal locality by changing the access pattern to data structures and processing multiple sequential patterns with a common prefix at a time to reduce the memory access latency by suppressing CPU cache misses. In this paper, we extend CC-PAID and dynamically reconstruct data structures, so that unnecessary data access to them can be avoided. The experimental results showed that CCDR-PAID executes up to 25% faster than CC-PAID because it sufficiently reduces cache misses and, therefore, provides better CPU cache utilization due to compaction of the data structure.