You are on page 1of 1

Association Rule Mining

work synergistically with the SOTrieIT and, through Agrawal, R., & Srikant, R. (1994). Fast algorithms
experiments with existing prominent algorithms and for mining association rules. Proceedings of the 20th
a variety of databases, we have proven the practicality International Conference on Very Large Databases
and superiority of our approach (Das, Ng, & Woon, (pp. 487-499), Chile.
2001; Woon et al., 2001). In fact, our latest algorithm,
Amir, A., Feldman, R., & Kashi, R. (1999). A new and
FOLD-growth, is shown to outperform FP-growth by
versatile method for association generation. Informa-
more than 100 times (Woon, Ng, & Lim, 2004).
tion Systems, 22(6), 333-347.
Babcock, B., Babu, S., Datar, M., Motwani, R., &
FUTURE TRENDS Widom, J. (2002). Models and issues in data stream
systems. Proceedings of the ACM SIGMOD/PODS
The data stream phenomenon will eventually become Conference (pp. 1-16), USA.
ubiquitous as Internet access and bandwidth become in-
Brijs, T., Swinnen, G., Vanhoof, K., & Wets, G. (1999).
creasingly affordable. With keen competition, products
Using association rules for product assortment deci-
will become more complex with customization and more
sions: A case study. Proceedings of the Fifth ACM
varied to cater to a broad customer base; transaction
SIGKDD Conference (pp. 254-260), USA.
databases will grow in both size and complexity. Hence,
association rule mining research will certainly continue Creighton, C., & Hanash, S. (2003). Mining gene ex-
to receive much attention in the quest for faster, more pression databases for association rules. Bioinformatics,
scalable and more configurable algorithms. 19(1), 79-86.
Das, A., Ng, W. K., & Woon, Y. K. (2001). Rapid as-
sociation rule mining. Proceedings of the 10th Inter-
CONCLUSION
national Conference on Information and Knowledge
Management (pp. 474-481), USA.
Association rule mining is an important data mining
task with several applications. However, to cope with Dong, G., & Li, J. (1999). Efficient mining of emerging
the current explosion of raw data, data structures must patterns: Discovering trends and differences. Proceed-
be utilized to enhance its efficiency. We have analyzed ings of the Fifth International Conference on Knowledge
several existing trie data structures used in association Discovery and Data Mining (pp. 43-52), USA.
rule mining and presented our novel trie structure, which
has been proven to be most useful and practical. What Elliott, K., Scionti, R., & Page, M. (2003). The conflu-
lies ahead is the parallelization of our structure to further ence of data mining and market research for smarter
accommodate the ever-increasing demands of today’s CRM. Retrieved from http://www.spss.com/home_
need for speed and scalability to obtain association rules page/wp133.htm
in a timely manner. Another challenge is to design new Han, J., Pei, J., Yin Y., & Mao, R. (2004). Mining fre-
data structures that facilitate the discovery of trends as quent patterns without candidate generation: A frequent-
association rules evolve over time. Different associa- pattern tree approach. Data Mining and Knowledge
tion rules may be mined at different time points and, by Discovery, 8(1), 53-97.
understanding the patterns of changing rules, additional
interesting knowledge may be discovered. Hidber, C. (1999). Online association rule mining.
Proceedings of the ACM SIGMOD Conference (pp.
145-154), USA.
REFERENCES Knuth, D.E. (1968). The art of computer programming,
Vol. 1. Fundamental Algorithms. Addison-Wesley
Aggarwal, C. C., & Yu, P. S. (2001). A new approach Publishing Company.
to online generation of association rules. IEEE Trans-
actions on Knowledge and Data Engineering, 13(4), Lawrence, R. D., Almasi, G. S., Kotlyar, V., Viveros, M.
527-540. S., & Duri, S. (2001). Personalization of supermarket

0

You might also like