A Novel Reduction Algorithm Based on Expert Knowledge

For complex systems and some new technology fields, simply rely on machines learning, the results are not reliable. Absorb the expert knowledge, will help us to more accurately grasp the status and the development of the complex or new fields. This paper presents a novel reduction algorithm that takes into account the knowledge of experts. The attributes set is divided into two subset according to the scores of the expert knowledge: the set of attributes with decisive expert knowledge and the set of attributes with experts' knowledge for reference, then  a novel man-machine cooperative intelligent reduction algorithm (IRAEK, Intelligent Reduction Algorithm based on Expert Knowledge) is proposed to find the minimum reduction based on the different scores of expert knowledge. Finally, the empirical analysis result on the Micro-electromechanical Systems (MEMS) field shows that the IRAEK algorithm is feasible and efficient.

[1]  Richard R. Reilly,et al.  An investigation of the validity of expert true score estimates in appraisal research , 1989 .

[2]  Anders Torvill Bjorvand.  Rough Enough - A System Supporting the Rough Sets Approach , 1997, SCAI.

[3]  Donghua Zhu,et al.  A New Reduction Algorithm Based on 0-1 Distinguish Matrix , 2006, 2006 6th World Congress on Intelligent Control and Automation.

[4]  T. L. Saaty.  A Scaling Method for Priorities in Hierarchical Structures , 1977 .

[5]  Yuan Junpeng,et al.  A Hierarchical Reduction Algorithm for Concept Hierarchy , 2006, Sixth International Conference on Intelligent Systems Design and Applications.

[6]  Andrzej Skowron,et al.  The Discernibility Matrices and Functions in Information Systems , 1992, Intelligent Decision Support.

[7]  Ron Kohavi,et al.  Supervised and Unsupervised Discretization of Continuous Features , 1995, ICML.

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

[9]  Yuan Junpeng,et al.  A Man-Machine Cooperative Intelligent Algorithm for Attribute Reduction , 2006, The Proceedings of the Multiconference on "Computational Engineering in Systems Applications".