An Interactive Rough Set Attribute Reduction Using Great Deluge Algorithm
Dimensionality reduction from an information system is a problem of eliminating unimportant attributes from the original set of attributes while avoiding loss of information in data mining process. In this process, a subset of attributes that is highly correlated with decision attributes is selected. In this paper, performance of the great deluge algorithm for rough set attribute reduction is investigated by comparing the method with other available approaches in the literature in terms of cardinality of obtained reducts (subsets), time required to obtain reducts, number of calculating dependency degree functions, number of rules generated by reducts, and the accuracy of the classification. An interactive interface is initially developed that user can easily select the parameters for reduction. This user interface is developed toward visual data mining.The carried out model has been tested on the standard datasets available in the UCI machine learning repository. Experimental results show the effectiveness of the method especially with relation to the time and accuracy of the classification using generated rules. The method outperformed other approaches in M-of-N, Exactly, and LED datasets with achieving 100% accuracy.
Keywordsinteractive data mining great deluge algorithm rough set theory attribute reduction classification
Unable to display preview. Download preview PDF.
- 1.Han, J., Kamber, M.: Data Mining: Concepts and Techniques. Morgan Kaufman Publishers, Oxford (2006)Google Scholar
- 6.Zhen, L., Xiangshi, R., Chaohai, Z.: User interface design of interactive data mining in parallel environment. In: Proceedings of the 2005 International Conference on Active Media Technology, AMT 2005, pp. 359–363 (2005)Google Scholar
- 11.Liu, H., Li, J., Wong, L.: A comparative study on feature selection and classification methods using gene expression profiles and proteomic patterns. In: Lathrop, R., Nakai, K., Miyano, S., Takagi, T., Kanehisa, M. (eds.) Genome Informatics 2002, vol. 13, pp. 51–60. Universal Academy Press, Tokyo (2002)Google Scholar
- 16.Jue, W., Hedar, A.R., Guihuan, Z., Shouyang, W.: Scatter Search for Rough Set Attribute Reduction. In: International Joint Conference on Computational Sciences and Optimization, CSO 2009, pp. 531–535 (2009)Google Scholar
- 17.Jensen, R., Shen, Q.: Finding Rough Set Reducts with Ant Colony Optimization. In: Workshop, U.K. (ed.) UK Workshop on Computational Intelligence, UK (2003)Google Scholar
- 19.Abdullah, S., Jaddi, N.S.: Great Deluge Algorithm for Rough Set Attribute Reduction. In: Zhang, Y., Cuzzocrea, A., Ma, J., Chung, K.-i., Arslan, T., Song, X. (eds.) DTA/BSBT 2010. CCIS, vol. 118, pp. 189–197. Springer, Heidelberg (2010)Google Scholar
- 20.Jaddi, N.S., Abdullah, S.: Nonlinear Great Deluge Algorithm for Rough Set Attribute Reduction. Journal of Information Science & Engineering 29, 49–62 (2013)Google Scholar
- 21.Mafarja, M., Abdullah, S.: Modified great deluge for attribute reduction in rough set theory. In: 2011 Eighth International Conference on Fuzzy Systems and Knowledge Discovery (FSKD), pp. 1464–1469 (2011)Google Scholar
- 22.Gunter, D.: New optimization heuristics (The Great Deluge Algorithm and Record to Record Travel). Computational Physic, 86–92 (1993)Google Scholar
- 23.Burke, E.K., Abdullah, S.: A Multi-start Large Neighbourhood Search Approach with Local Search Methods for Examination Timetabling. In: Long, D., Smith, S., Borrajo, D., McCluskey, L. (eds.) Proceedings of the International Conference on Automated Planning and Scheduling (ICAPS 2006), Cumbria, UK (2006)Google Scholar
- 27.Migut, M., Worring, M.: Visual exploration of classification models for various data types in risk assessment. Information Visualization (2012)Google Scholar