Missing Values Imputation Based on Iterative Learning

Huaxiong Li

School of Management and Engineering, Nanjing University, Nanjing, China.

DOI: 10.4236/ijis.2013.31A006                

Abstract

Databases for machine learning and data mining often have missing values. How to develop effective method for missing values imputation is a crucial important problem in the field of machine learning and data mining. In this paper, several methods for dealing with missing values in incomplete data are reviewed, and a new method for missing values imputation based on iterative learning is proposed. The proposed method is based on a basic assumption: There exist cause-effect connections among condition attribute values, and the missing values can be induced from known values. In the process of missing values imputation, a part of missing values are filled in at first and converted to known values, which are used for the next step of missing values imputation. The iterative learning process will go on until an incomplete data is entirely converted to a complete data. The paper also presents an example to illustrate the framework of iterative learning for missing values imputation.

Keywords

Incomplete Data; Missing Values Imputation; Iterative Learning; Intension; Extension

Conflicts of Interest

The authors declare no conflicts of interest.

References

[1]	T. M. Mitchell, “Generalization as Search,” Artificial Intelligence, Vol. 18, No. 2, 1982, pp. 203-226. doi:10.1016/0004-3702(82)90040-6
[2]	T. M. Mitchell, “Machine Learning,” McGraw-Hill, New York, 1997.
[3]	J. R. Quinlan, “C4.5: Programs for Machine Learning,” Morgan Kaufmann, San Mateo, 1993.
[4]	P. Clark and T. Niblett, “The CN2 Induction Algorithm,” Machine Learning, Vol. 3, No. 4, 1989, pp. 261-283. doi:10.1007/BF00116835
[5]	J. W. Grzymala-Busse and W. J. Grzymala-Busse, “An Experimental Comparison of Three Rough Set Approaches to Missing Attribute Values,” In: J. Peters, A. Skowron, I. Duntsch, J. Grzymala-Busse, E. Orlowska and L. Polkowski, Eds. LNCS Transactions on Rough Sets VI, Springer, Berlin, 2007, pp. 31-50.
[6]	J. W. Grzymala-Busse, “On the Unknown Attribute Values in Learning from Examples,” In: Z. Ras and M. Zemankova, Eds., Proceedings of 6th International Symposium on Methodologies for Intelligent Systems, Springer, Berlin, 1991, pp. 368-377.
[7]	Z. Ghahramani and M. I. Jordan, “Supervised Learning from Incomplete Data via an EM Approach,” In: J. D. Cowan, G. Tesauro and J. Alspector, Eds., Advances in Neural Information Processing Systems, Morgan Kaufmann, San Mateo, 1994, pp. 120-127.
[8]	S. Greco, B. Matarazzo and R. Slowinski, “Handling Missing Values in Rough Set Analysis of Multi-Attribute and Multi-Criteria Decision Problems,” In: N. Zhong, A. Skowron and S. Ohsuga, Eds., Proceedings of 7th International Workshop on Rough Sets, Fuzzy Sets, Data Mining, and Granular-Soft Computing, Springer, Berlin, 1999, pp. 146-157.
[9]	M. Kryszkiewicz, “Rough Set Approach to Incomplete Information Systems, Information Sciences, Vol. 112, No. 1-4, 1998, pp. 39-49. doi:10.1016/S0020-0255(98)10019-1
[10]	M. Kryszkiewicz, “Rules in Incomplete Information Systems,” Information Sciences, Vol. 113, No. 3-4, 1999, pp. 271-292. doi:10.1016/S0020-0255(98)10065-8
[11]	J. Stefanowski and A. Tsoukiàs, “On the Extension of Rough Sets under Incomplete Information,” International Journal of Intelligent System, Vol. 16, No. 1, 2000, pp. 29-38.
[12]	H. X. Li, Y. Y. Yao, X. Z. Zhou and B. Huang, “Two-Phase Rule Induction from Incomplete Data,” In: G. Wang, T. Li, J. Grzymala-Busse, D. Miao, A. Skowron and Y. Yao, Eds., Proceedings of 3rd International Conference on Rough Sets and Knowledge Technology, Springer, Berlin, pp. 47-54.
[13]	B. Shahshahani and D. Landgrebe, “The Effect of Unlabeled Samples in Reducing the Small Sample Size Problem and Mitigating the Hughes Phenomenon,” IEEE Transactions on Geoscience and Remote Sensing, Vol. 32, No. 5, 1994, pp. 1087-1095. doi:10.1109/36.312897
[14]	Y. Y. Yao, “Concept Formation and Learning: A Cognitive Informatics Perspective,” In: C. Chan, W. Kinsner, Y. Wang and D. Miller, Eds., Proceedings of 3rd IEEE International Conference on Cognitive Informatics, IEEE CS Press, New York, 2004, pp. 42-51.
[15]	Y. Y. Yao and N. Zhong, “An Analysis of Quantitative Measures Associated with Rules,” In: X. Wu, K. Ramamohanarao and K. Korb, Eds., Proceedings of 2nd Pacific-Asia Conference on Knowledge Discovery and Data Mining, Springer, Berlin, 1999, pp. 479-488.
[16]	Y. Zhao, Y. Y. Yao and J. T. Yao, “Level-Wise Construction of Decision Trees for Classification,” International Journal of Software Engineering and Knowledge Engineering, Vol. 16, No. 1, 2006, pp. 103-123. doi:10.1142/S0218194006002690
[17]	J. T. Yao and Y. Y. Yao, “Induction of Classification Rules by Granular Computing,” In: J. Alpigini, J. Peters, A. Skowron and N. Zhong, Eds., Proceedings of 3rd International Conference on Rough Sets and Current Trends in Computing, Springer, Berlin, 2002, pp. 331-338. doi:10.1007/3-540-45813-1_43