Mining Indirect Temporal Sequential Patterns in Large Transaction Databases

Abstract:

Sequential pattern is an important research topic in data mining and knowledge discovery. Traditional algorithms for mining sequential patterns focus on the frequent sequences, which do not consider the infrequent sequences and lifespan of each sequence. On the one hand, some infrequent patterns can provide very useful insight view into the data set, on the other hand, without taking lifespan of each sequence into account, not only some discovered patterns may be invalid, but also some useful patterns may not be discovered. So, we extend the sequential patterns to the indirect temporal sequential patterns, and put forward an algorithm to discover indirect temporal sequential patterns in this paper.

• Periodical: Applied Mechanics and Materials (Volumes 385-386)
• Pages: 1362-1365
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.385-386.1362
• Online since: August 2013
• Authors: Wei Min Ouyang, Qin Hua Huang
• Keywords: Pattern, Sequence, Temporal

References

[1] Agrawal R, Srikant R. Fast Algorithms for Mining Association Rules In the Proc. of the 20th International Conference on VLDB. Santiago, 1994. p.487~499.

[2] Agrawal R, Srikant R. Mining sequential patterns. In the Proc. 1995 Int Conf. on Data Engineering, Taibei, Taiwan, March 1995, pp.3-14.

[3] P.N. Tan and V. Kumar. Indirect Association: Mining Higher Order Dependences in Data. Proc. Of the 4th European Conference on Principles and Practice of Knowledge Discovery in Databases, pp.632-737, Lyon, France(2000).

DOI: 10.1007/3-540-45372-5_77

[4] H. Cao, N. Mamoulis, and D.W. Cheung, Mining Frequent Spatio-Temporal Sequential Patterns, " Proc. Fifth IEEE Int'l Conf. Data Mining (ICDM , 05), pp.82-89, (2005).

DOI: 10.1109/icdm.2005.95