AUTHORS: Hyontai Sug

Download as PDF

ABSTRACT: Inductive learning algorithms want to find a function that reflects a given sequence of input and output pairs where the input and output pairs consist of value vectors. Rough set systems can extract minimal set of rules that act as the function of inductive learning, and are well known for their strong mathematical background. The found set of rules is solely based on data so that no prejudiced views can be inserted in the found rule set. But even though the good property, rough set based rule systems have the tendency of being unstable in the sense that their performance is very dependent on given training data sets due to their sole reliance on given data. In order to avoid such property of the rough set based rule systems this paper suggests using the attributes of frequent items only in the input vector to find the rules. The attributes can be found by applying association rule algorithms. Experiments with several real world data sets show that better rough set based rules could be found in accuracy by using the attributes only, especially when the attributes of input have key-like characteristics.

KEYWORDS: Inductive learning, rough set theory, association rules, frequent attributes, data mining

REFERENCES:

[1] Z. Pawlak, Rough Sets, International Journal of Parallel Programming, Vol. 15, No. 5, 1982, pp. 341-356.

[2] F. Fleuret, P. Abbet, C. Dubout, L. Lefakis, The MASH Project, Lectures Notes in Computer Science, Vol. 6913, 2011, pp. 626- 629.

[3] M. Bal, Rough Sets Theory as Symbolic Data Mining method: An Application on Complete Decision Table, Information Science Letters, Vol. 2, No. 1, 2013, pp. 35-47.

[4] R. Agrawal, T. Imieliński, A. Swami, Mining association rules between sets of items in large databases, Proceedings of the 1993 ACM SIGMOD international conference on Management of data - SIGMOD '93, 1993, pp. 207-216.

[5] J. Hipp, U. Guntzer, G. Nakhaeizadeh, Algorithms for association rule mining – a general survey and comparison, ACM SIGKDD Explorations Newsletter, Vol. 2, No. 58, 2000, pp. 58-64

[6] J. Li, H. Shen, R. Toper, Mining the optimal class association rule set, Knowledge-Based Systems, Vol. 15, 2002, pp. 399-405.

[7] P. Deekumpa, P. Sooraksa, Associate rule minimization using Boolean algebra set function, Proceedings of 4th International Joint Conference on Information and Communication Technology, Electronic and Electrical Engineering (JICTEE2014), 2014, DOI: 10.1109/JICTEE.2014.6804099.

[8] E.F. Ashmouni, R.A. Ramadan, A.A. Rashed, Espresso for Rule Mining, Procedia Computer Science, Vol. 32, 2014, pp. 596-603.

[9] H. Sug, Using Machine Learning Methods Jointly to Find Better Set of Rules in Data Mining, Proceedings of 21st International Conference on Circuits, Systems, Communications, and Computers - CSCC 2017, July 2017.

[10] J. Stefanowski, The rough set based rule induction technique for classification problems. Proceedings of the 6th European Conference on Intelligent Techniques and Soft Computing, EUFIT-98, 1998, pp. 109-113.

[11] L. Breiman, Bagging predictors, Machine Learning, Vol. 24, No. 2, 1996, pp. 123-140.

[12] J. Stefanowski, The bagging and n2-classifiers based on rules induced by MODLEM, Lecture Notes in Artificial Intelligence, Vol. 3066, Springer-Verlag, 2004, pp. 488-497.

[13] J. Stefanowski, On Combined Classifiers, Rule Induction and Rough Sets, Lecture Notes of Computer Science, Vol. 4374, Splinger-Verlag, 2007, pp. 329-350.

[14] T. G. Dietterich, An Experimental Comparison of Three Methods for Constructing Ensembles of Decision Trees: Bagging, Boosting, and Randomization, Machine Learning, Vol. 40, Issue 2, 2000, pp. 139-157.

[15] Quinlan, J.R., C4.5: Programs for Machine Learning, Morgan Kaufmann Publishers, Inc., 1993.

[16] L. Breiman, J. Friedman, R. Olshen, C. Stone, Classification and Regression Trees, Wadsworth International Group, 1984.

[17] J. Stefanowski, S. Wilk, Improving Rule-Based Classifiers Induced by MODLEM by Selective Pre-processing of Imbalanced Data, Proceedings of the International Workshop on Rough Sets in Knowledge Discovery: Foundations and Applications, ECML 2007 – RSKD’07, 2007, pp. 54-65.

[18] T. Slimani, Class Association Rules Mining based Rough Set Method, International Journal of Engineering and Technology, Vol. 6, No. 6, 2015, pp. 2786-2794.

[19] S. Liao, H. Chang, A rough set-based association rule approach for a recommendation system for online customers, Information Processing & Management, Vol. 52, Issue 6, pp. 1142-1160, 2016.

[20] S.B. Sagar, A. Tiwari, Rough set and genetic based approach for maximization of weighted association rules, International Journal of Education and Computer Science, Vol. 8, No. 3, pp. 54-63, 2016.

[21] D.J. Leinweber, Stupid Data Miner tricks, The Journal of Investing, Vol. 16, No. 1, pp. 15-22. 2007.

[22] J. Brownlee, Overfitting and Underfitting with Machine Learning Algorithms, http://machinelearningmastery.com/overfittingand-underfitting-with-machine-learningalgorithms/, 2016.

[23] A. Frank, A. Suncion, UCI Machine Learning Repository

[http://archive.ics.uci.edu/ml]. Irvine, CA: University of California, School of Information and Computer Sciences, 2010.

[24] J. Stefanowski, The rough set based rule induction technique for classification problems, Proceedings of 6th European Congress on Intelligent Techniques and Soft Computing, pp. 109-113 1998.

[25] U. M. Fayyad, K.B. Irani, Multi-Interval Discretization of Continuous-Valued Attributes for Classification Learning, Proceedings of Thirteenth International Joint Conference on Artificial Intelligence, pp. 1022-1027, 1993.