A Feature Selection Approach of Inconsistent Decision Systems in Rough Set

Feature selection has been widely discussed as an important preprocessing step in data mining applications since it reduces a model's complexity. In this paper, limitations of several representative reduction methods are analyzed firstly, and then by distinguishing consistent objects form inconsistent objects, decision inclusion degree and its probability distribution function as a new measure are presented for both inconsistent and consistent simplified decision systems. New definitions of distribution reduct and maximum distribution reduct for simplified decision systems are proposed. Many important propositions, properties, and conclusions for reduct are drawn. By using radix sorting and hash techniques, a heuristic distribution reduct algorithm for feature selection is constructed. Finally, compared with other feature selection algorithms on six UCI datasets, the proposed approach is effective and suitable for both consistent and inconsistent decision systems.

[1]  Dai Chun-yan,et al.  A survey on rough set theory and its application , 2004 .

[2]  Lin Sun,et al.  A New Knowledge Reduction Algorithm Based on Decision Power in Rough Set , 2010, Trans. Rough Sets.

[3]  Xu Zhang,et al.  A Quick Attribute Reduction Algorithm with Complexity of max(O(｜C｜｜U｜),O(｜C｜~2｜U/C｜)) , 2006 .

[4]  Lin Sun,et al.  Knowledge Entropy and Feature Selection in Incomplete Decision Systems , 2013 .

[5]  Lin Sun,et al.  Feature selection using mutual information based uncertainty measures for tumor classification. , 2014, Bio-medical materials and engineering.

[6]  Lin Sun,et al.  Feature selection using rough entropy-based uncertainty measures in incomplete decision systems , 2012, Knowl. Based Syst..

[7]  Lin Sun,et al.  Granularity Partition-based Feature Selection and Its Application in Decision Systems , 2012 .

[8]  Lin Sun,et al.  Decision Degree-based Decision Tree Technology for Rule Extraction , 2012, J. Comput..

[9]  Lin Sun,et al.  Information Quantity-based Decision Rule Acquisition from Decision Tables , 2012 .

[10]  Malcolm J. Beynon,et al.  Reducts within the variable precision rough sets model: A further investigation , 2001, Eur. J. Oper. Res..

[11]  Lin Sun,et al.  Granular Computing-based Granular Structure Model and its Application in Knowledge Retrieval , 2012 .

[12]  Fan Min,et al.  Knowledge Reduction in Inconsistent Decision Tables , 2006, ADMA.

[13]  Janusz Zalewski,et al.  Rough sets: Theoretical aspects of reasoning about data , 1996 .

[14]  Wei-Zhi Wu,et al.  Knowledge reduction in random information systems via Dempster-Shafer theory of evidence , 2005, Inf. Sci..

[15]  Lin Sun,et al.  Approaches to Knowledge Reduction of Decision Systems based on Conditional Rough Entropy , 2011 .

[16]  Marzena Kryszkiewicz.  Comparative study of alternative types of knowledge reduction in inconsistent systems , 2001, Int. J. Intell. Syst..

[17]  Jun Fang,et al.  A fast feature selection approach based on rough set boundary regions , 2014, Pattern Recognit. Lett..

[18]  Dongyi Ye,et al.  A Novel Maximum Distribution Reduction Algorithm for Inconsistent Decision Tables , 2006, KSEM.

[19]  Marzena Kryszkiewicz.  Comparative study of alternative types of knowledge reduction in inconsistent systems , 2001, Int. J. Intell. Syst..

[20]  Zheng Pei,et al.  The Relationship Among Several Knowledge Reduction Approaches , 2005, FSKD.

[21]  Yiyu Yao,et al.  A Survey on Rough Set Theory and Applications: A Survey on Rough Set Theory and Applications , 2009 .

[22]  Wei-Zhi Wu,et al.  Approaches to knowledge reductions in inconsistent systems , 2003, Int. J. Intell. Syst..

[23]  Chu Jian.  Quick Attribute Reduction Algorithm with Hash , 2009 .

[24]  Bo Zhang,et al.  The Quotient Space Theory of Problem Solving , 2003, Fundam. Informaticae.

[25]  Lin Sun,et al.  Granular Space-Based Feature Selection and Its Applications , 2013, J. Softw..

[26]  Miao Duo,et al.  A HEURISTIC ALGORITHM FOR REDUCTION OF KNOWLEDGE , 1999 .

[27]  Wang Guo,et al.  Decision Table Reduction based on Conditional Information Entropy , 2002 .

[28]  Wei-Zhi Wu,et al.  Approaches to knowledge reduction based on variable precision rough set model , 2004, Inf. Sci..

[29]  Liu Shao,et al.  Research on Efficient Algorithms for Rough Set Methods , 2003 .

[30]  Lin Sun,et al.  A granular computing approach to gene selection. , 2014, Bio-medical materials and engineering.

[31]  Lin Sun,et al.  Rough Entropy-based Feature Selection and Its Application ⋆ , 2011 .

[32]  Yiyu Yao,et al.  Attribute reduction in decision-theoretic rough set models , 2008, Inf. Sci..