Abstract

We are given a large database of customer transactions, where each transaction consists of transaction-id, the items bought in the transaction and the transaction time. The whole set of transaction is divided into a number of segments called durations (intervals) based on transaction time. And the dividing standard can be monthly, quarterly or yearly. We introduce the problem of mining strong association rules between consecutive durations using FP-tree and correlation coefficient, which is used to quantitatively describe the strength and sign of a relationship between two variables. This paper deals with the changes in the correlation between any two itemsets at the transition of the consecutive duration. Milestone is a change over point between durations. The transition may be positive or negative which are time points at which the pattern is either positively or negatively correlated. Also the method provides rare items, whose support is poor but are highly correlated.

References

• R. Agrawal, T. Imielinski, and A. Swami, “Mining Association Rules between Sets of Items in Large Databases,” Proc. 1993 ACM SIGMOD Int’l Conf. Management of Data (SIGMOD ’93), pp. 207-216, 1993.
• J.Han, J. Pei and Y. Yin. “Mining frequent patterns without candidate generation”. In ACM SIGMOD 2000.
• Wen-Chi Hou.“Extraction and applications of statistical relationships in relational databases.”. IEEE Transactions on knowledge and data engineering, 1996.
• H. Xiong, S. Shekhal, P. N. Tan, V. Kumar, “Exploiting a support-based Upper Bound of Pearson’s Correlation Coefficient for Efficiently Identifying stringly correlated pairs.
• J. Zhang, J. Feigenbaum, “Finding highly correlated pairs efficiently with powerful pruning.”
• Q. Wan and A. An, “Transitional patterns and their significant milestones.” Proc. Seventh IEEE International conference Data Mining,2007.
• S. Brin, R. Motwani, and C. Silverstein. “Beyond market baskets: Generalizing association rules to correlations.”. In ACM SIGMOD, 1997.
• Wu, X. Zhang, C. Zhang, “ Mining both positive and negative association rules.”. In Proc. ICML, 2002.
• Z. Zhang, Z. Chen, “Time series data mining method based on lightly-supported boolean association rules.” Proc. CIKM’06.
• G.-Z. Dong and J.-Y. Li, “Efficient Mining of Emerging Patterns: Discovering Trends and Differences,” Proc. Fifth ACM SIGKDD Int’l Conf. Knowledge Discovery and Data Mining (KDD ’99), pp. 43-52, 1999.
• B. O ¨ zden, S. Ramaswamy, and A. Silberschatz, “Cyclic Association Rules,” Proc. 14th Int’l Conf. Data Eng. (ICDE ’98), pp. 412-421, 1998.
• J.M. Ale and G.H. Rossi, “An Approach to Discovering Temporal Association Rules,” Proc. 2000 ACM Symp. Applied Computing (SAC’00), pp. 294-300, 2000.
• R. Agrawal and R. Srikant, “Mining Sequential Patterns,” Proc. 11th Int’l Conf. Data Eng., pp. 3-14, 1995.
• R. Agrawal and R. Srikant, “Mining Sequential Patterns: Generalizations and Performance Improvements,” Proc. Fifth Int’l Conf. Extending Database Technology (EDBT ’96), 1996.
• J. Pei, J. Han, B. Mortazavi-Asl, J.-Y. Wang, H. Pinto, Q.-M. Chen,U. Dayal, and M.-C. Hsu, “Mining Sequential Patterns by Pattern-Growth: The Prefixspan Approach,” IEEE Trans. Knowledge and Data Eng., vol. 16, no. 11, pp. 1424-1440, Nov. 2004.