Abstract
In this paper, an Enhanced Class-Attribute Interdependence Maximization discretization algorithm (ECAIM) is proposed by 2 extensions to improve a state-of-the-art Class-Attribute Interdependence Maximization discretization algorithm (CAIM). The main drawback that remains unresolved in CAIM is that its stopping criterion depends on the number of target classes. When the number of target classes is large, its performance drops, as CAIM is not a real incremental discretization method. The first extension, ECAIM is extended from CAIM to become a real incremental discretization method by improving the stopping criterion. The stopping criterion is based on the Slope of an ecaim value which decreases with an increasing number of intervals. If the slope of ecaim value is less than the specified threshold then the discretization terminates. The second extension that we propose is the multi-attribute techniques by simultaneously considering all attributes instead of a single-attribute like CAIM, for accurate and efficient discretizers solution. ECAIM use a feature selection algorithm to select a subset of attributes for reducing the number of attributes, remove irrelevant, redundant attributes and then use multi-attribute techniques only on this subset attributes. Experiment results on 15 real-world datasets show that ECAIM is more efficient than CAIM in terms of accuracy, number of intervals and number of generated rules.
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
Kurgan, L.A., Cios, K.J.: CAIM discretization algorithm. IEEE Transactions on Knowledge and Data Engineering, 145–153 (2004)
Tsai, C.-J., Lee, C.-I., Yang, W.-P.: A discretization algorithm based on Class-Attribute Contingency Coefficient. Information Sciences 178, 714–731 (2008)
Kerber, R.: ChiMerge: discretization of numeric attributes. In: Proceedings of the Tenth National Conference on Artificial Intelligence, pp. 123–128. AAAI Press, San Jose (1992)
Ruiz, F.J., Angulo, C., Agell, N.: IDD: A Supervised Interval Distance-Based Method for Discretization. IEEE Transactions on Knowledge and Data Engineering 20, 1230–1238 (2008)
GarcÃa, S., Luengo, J., Saez, J., Lopez, V., Herrera, F.: A Survey of Discretization Techniques: Taxonomy and Empirical Analysis in Supervised Learning. IEEE Transactions on Knowledge and Data Engineering, 1 (2012)
Liu, H., Hussain, F., Tan, C.L., Dash, M.: Discretization: An Enabling Technique. Data Mining and Knowledge Discovery 6, 393–423 (2002)
Vyas, O.P., Das, K.: Article: A Suitability Study of Discretization Methods for Associative Classifiers. International Journal of Computer Applications 5, 46–51 (2010)
Dougherty, J., Kohavi, R., Sahami, M.: Supervised and Unsupervised Discretization of Continuous Features. In: Prieditis, A., Russell, S. (eds.) Proceedings of the Twelfth International Conference on Machine Learning, Tahoe City, California, USA, pp. 194–202 (1995)
Huan, L., Setiono, R.: Feature selection via discretization. IEEE Transactions on Knowledge and Data Engineering 9, 642–645 (1997)
Zhu, Q., Lin, L., Shyu, M.-L., Chen, S.-C.: Effective supervised discretization for classification based on correlation maximization. In: 2011 IEEE International Conference on Information Reuse and Integration (IRI), pp. 390–395 (2011)
GarcÃa, M.N.M., Lucas, J.P., Batista, V.F.L., MartÃn, M.J.P.: Multivariate Discretization for Associative Classification in a Sparse Data Application Domain. In: Graña Romay, M., Corchado, E., Garcia Sebastian, M.T. (eds.) HAIS 2010, Part I. LNCS, vol. 6076, pp. 104–111. Springer, Heidelberg (2010)
Huan, L., Setiono, R.: A Probabilistic Approach to Feature Selection - A Filter Solution. In: ICML, pp. 319–327 (1996)
Frank, A., Asuncion, A.: Machine Learning Repository (2010)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Sriwanna, K., Puntumapon, K., Waiyamai, K. (2012). An Enhanced Class-Attribute Interdependence Maximization Discretization Algorithm. In: Zhou, S., Zhang, S., Karypis, G. (eds) Advanced Data Mining and Applications. ADMA 2012. Lecture Notes in Computer Science(), vol 7713. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-35527-1_39
Download citation
DOI: https://doi.org/10.1007/978-3-642-35527-1_39
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-35526-4
Online ISBN: 978-3-642-35527-1
eBook Packages: Computer ScienceComputer Science (R0)