Abstract
Emerging patterns are itemsets whose frequencies change sharply from one class to the other. PCL is an example of efficient classification algorithms that leverage the prediction power of emerging patterns. It first selects the top-K emerging patterns of each class that match a testing instance, and then uses these selected patterns to decide the class label of the testing instance. We study the impact of the parameter K on the accuracy of PCL. We have observed that in many cases, the value of K is critical to the performance of PCL. This motivates us to develop an algorithm to find the best value of K for PCL. Our results show that finding the best K can improve the accuracy of PCL greatly, and employing incremental frequent itemset maintenance techniques reduces the running time of our algorithm significantly.
Supported by A*STAR grant (SERC 072 101 0016).
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Bailey, J., Manoukian, T., Ramamohanarao, K.: Classification using constrained emerging patterns. In: Dong, G., Tang, C., Wang, W. (eds.) WAIM 2003. LNCS, vol. 2762, pp. 226–237. Springer, Heidelberg (2003)
Cheng, H., et al.: Discriminative frequent pattern analysis for effective classification. In: Proc 23th ICDE, pp. 716–725 (2007)
Cheng, H., et al.: Direct discriminative pattern mining for effective classification. In: Proc 24th ICDE, pp. 169–178 (2008)
Dong, G., et al.: CAEP: Classification by Aggregating Emerging Patterns. In: Proc. 2nd Intl. Conf. on Discovery Science, pp. 30–42 (1999)
Feng, M.: Frequent Pattern Maintenance: Theories and Algorithms. PhD thesis, Nanyang Technological University (2009)
Feng, M., et al.: Negative generator border for effective pattern maintenance. In: Proc 4th Intl. Conf. on Advanced Data Mining and Applications, pp. 217–228 (2008)
Feng, M., et al.: Evolution and maintenance of frequent pattern space when transactions are removed. In: Zhou, Z.-H., Li, H., Yang, Q. (eds.) PAKDD 2007. LNCS (LNAI), vol. 4426, pp. 489–497. Springer, Heidelberg (2007)
Li, W., Han, J., Pei, J.: CMAR: Accurate and efficient classification based on multiple class-association rules. In: Proc 1st ICDM, pp. 369–376 (2001)
Li, J., Wong, L.: Solving the fragmentation problem of decision trees by discovering boundary emerging patterns. In: Proc. 2nd ICDM, pp. 653–656 (2002)
Li, J., Wong, L.: Structural geography of the space of emerging patterns. Intelligent Data Analysis 9(6), 567–588 (2005)
Li, J., Ramamohanarao, K., Dong, G.: The space of jumping emerging patterns and its incremental maintenance algorithms. In: Proc. of 17th ICML, pp. 551–558 (2000)
Li, J., et al.: Minimum description length principle: Generators are preferable to closed patterns. In: Proc. 21st Natl. Conf. on Artificial Intelligence, pp. 409–415 (2006)
Li, H., et al.: Relative risk and odds ratio: A data mining perspective. In: Proc 24th PODS, pp. 368–377 (2005)
Liu, B., Hsu, W., Ma, Y.: Integrating classification and association rule mining. In: Proc. 4th KDD, pp. 80–86 (1998)
Ramamohanarao, K., Fan, H.: Patterns based classifiers. In: Proc. 16th WWW, pp. 71–83 (2007)
Ramamohanarao, K., Bailey, J.: Discovery of emerging patterns and their use in classification. In: Gedeon, T(T.) D., Fung, L.C.C. (eds.) AI 2003. LNCS (LNAI), vol. 2903, pp. 1–12. Springer, Heidelberg (2003)
Thabtah, F.A., Cowling, P., Peng, Y.: MMAC: A new Multi-class Multi-label Associative Classification approach. In: Proc. 4th ICDM, pp. 217–224 (2004)
Yin, X., Han, J.: CPAR: Classification based on Predictive Association Rules. In: Proc. 3rd SDM, pp. 331–335 (2003)
Wang, J., Karypis, G.: HARMONY: Efficiently mining the best rules for classification. In: Proc. 5th SDM, pp. 205–216 (2005)
Zhang, X., Dong, G., Ramamohanarao, K.: Information-based classification by aggregating emerging patterns. In: Proc. 2nd Intl. Conf. on Intelligent Data Engineering and Automated Learning, pp. 48–53 (2000)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2010 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Ngo, TS., Feng, M., Liu, G., Wong, L. (2010). Efficiently Finding the Best Parameter for the Emerging Pattern-Based Classifier PCL. In: Zaki, M.J., Yu, J.X., Ravindran, B., Pudi, V. (eds) Advances in Knowledge Discovery and Data Mining. PAKDD 2010. Lecture Notes in Computer Science(), vol 6118. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-13657-3_15
Download citation
DOI: https://doi.org/10.1007/978-3-642-13657-3_15
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-13656-6
Online ISBN: 978-3-642-13657-3
eBook Packages: Computer ScienceComputer Science (R0)