A Rough Set Classifier Based on Discretization and Attribute Selection
Rough set is useful for classification in learning model. The main advantage of rough set is to find out the relativity among attributes directly from attributes of data without any preliminary or additional information. A rough set classifier based on discretization and attribute selection is proposed in this paper. Our rough set classifying algorithm give a full consider about condition attribute significance during the process of rule forming. We verified our algorithm on five well-known UCI machine learning data sets. The experiment results are expressed by mean of accuracies. At last, we compare our experiment results with the classical algorithms of other two references  and . Results prove proposed algorithm is better than them. It can get higher classification accuracy, lower breakpoints and rules in all data sets of our experiments.
KeywordsClassification Discretization Interval division Attribute significance 10-fold cross validation
This work is supported by the Project of Research on Science and Technology of Jilin Education Ministry of China under Grant No. 2014249, No. 2015367 and No. 2013250, the Special Project of Jilin Province Industrial Technology Research and Development of China under Grant No. 2014Y101, No. 2019C052, the Research Foundation Project of Changchun normal University of China under Grant No. 2017015 and the financial support from the program of China Scholarship Council, No. 201408220056.
- 6.Dougherty, J., Kohavi, R., Sahami, M.: Supervised and unsupervised discretization of continuous features. In: Proceedings of the 12th International Conference on Machine Learning, pp. 194–202. Morgan Kaufmann Publishers (1995)Google Scholar
- 7.Jefferys, W.H., Berger, J.O.: Ockham’s razor and Bayesian analysis. Am. Sci. 80(1), 64–72 (1992)Google Scholar
- 11.Shi, Z., Xia, Y., Wu, F., Dai, J.: The discretization algorithm for rough data and its application to intrusion detection. J. Netw. 9(6), 1380 (2014)Google Scholar