Regular-frequent patterns are an important type of regularities that exist in transactional, time-series and any other types of databases. A frequent pattern can be said regular-frequent if it appears at a regular interval given by the user specified threshold in the transactional database. The regularity calculation for every candidate pattern is a computationally expensive process, especially when there exist long patterns. Currently the FP-growth algorithm is one of the most popular and fastest approaches to mining periodic frequent item sets. Therefore, in this paper we introduce a novel concept of mining regular-frequent patterns (RFP) in transactional databases. We introduce two mining techniques based on transaction number and also based on products or itemsets on the vertical data format. The efficiency is achieved by eliminating aperiodic or irregular patterns during execution based on suboptimal solutions. Our tree based structure helps to captures the database contents in highly compact manner. Our experimental results are highly efficient and scalable as well as improve the overall response time.

Frequent patterns, regular patterns, transactional databases, vertical data format

[1] R. Agrawal, T. Imielinski, A.N. Swami “Mining Association Rules Between Sets of Items in Large Databases”, International Conference on Management of Data(ACM SIGMOD), pp. 207–216, 1993.
[2] R.U. Kiran, P.K. Reddy “Towards efficient mining of periodic frequent patterns in transactional databases”, DEXA, pp. 194-208, 2010.
[3] J. Han, J. Pei, Y.Yin “Mining Frequent Patterns without Candidate Generation”, International Conference on Management of Data(ACM SIGMOD), pp. 1–12, 2000.
[4] K. Amphawan, P. Lenca, A. Surarerks “Mining top-k periodicfrequent pattern from transactional databases without support threshold”, Advances in Information Technology, pp. 18–29, 2009.
[5] J.N. Venkatesh, R.U. Kiran, K, Reddy, M. Kitsuregawa “Discovering Periodic-Frequent Patterns in Transactional Databases Using All-Confidence and Periodic-All-Confidence”, Springer International Publishing,Switzerland, pp. 55-70, 2016.
[6] S.K. Tanbeer, C.F. Ahmed, B.-S. Jeong, Y.-K. Lee, “Discovering periodic frequent patterns in transactional databases” Theeramunkong, T., Kijsirikul, B., Cercone, N., Ho, T.-B. (eds.) , Springer, Heidelberg, PAKDD, LNCS, Vol. 5476, pp. 242–253,2009.
[7] J. Pei, J. Han “Constrained Frequent Pattern Mining: A Pattern-Growth View*”, SIGKDD Explorations, Vol.4, Issue.1, pp.31-39.
[8] R.U. Kiran, M. Kitsuregawa, P.K. Reddy, “Efficient Discovery of Periodic-Frequent Patterns in Very Large Databases” Journal of Systems and Software, 2015.
[9] A. Surana, R.U. Kiran, P.K. Reddy “An efficient approach to mine periodic frequent patterns in transactional databases” Cao, L., Huang, J.Z., Bailey, J., Koh, Y.S., Luo, J. (eds.) PAKDD Workshops 2011, Springer, Heidelberg (2012), LNCS, Vol.7104, pp.254–266, 2012.
[10] V.M. Nofong “Discovering Productive Periodic Frequent Patterns in Transactional Databases”, Springer-Verlag Berlin Heidelberg, 2016.
[11] R.U. Kiran, M. Kitsuregawa “Discovering quasi-periodic-frequent patterns in transactional databases” Bhatnagar V, Srinivasa S (eds) BDA 2013, LNCS, Springer International Publishing, Heidelberg, pp97–115, 2013.
[12] V. Kumar, V. Kumari “Incremental mining for regular frequent patterns in vertical format” International Journal of Engineering Technology, Vol.5, Issue.2, pp.1506–1511, 2013.
[13] M.J. Zaki “Parallel and distributed association mining: A survey” IEEE concurrency, pp. 14-25, 1999.
[14] T. Hu, S.Y. Sung, H. Xiong, Q. Fu “Discovery of Maximum Length Frequent Itemsets” Information Sciences, pp.69–87, 2008.
[15] N.Sethi, P.Sharma “Mining Frequent Pattern from lLarge Dynamic Database Using Compacting Data Sets” International Journal of Scientific Research in Computer Science and Engineering, Vol.1, Issue.3, pp.31–34, 2013.