期刊名称：International Journal of Computer Science and Network Security
印刷版ISSN：1738-7906
出版年度：2019
卷号：19
期号：10
页码：165-171
出版社：International Journal of Computer Science and Network Security
摘要：Attribute reduction problem can be defined as a process of eliminating redundant attributes, while avoiding information loss. It is known to be an NP-hard optimization problem, which deals with finding the minimal attribute from a large set of attributes. Many heuristic and meta-heuristic approaches have been widely used by researchers. However, this study will focus on the use of a reinitiate great deluge algorithm with a composite neighbourhood structure (rGD-cNbs), proposed for the rough set of attribute reduction problem. rGD-cNbs is a meta-heuristic approach, which is based on a basic great deluge. Its difference is that the level of the great deluge is reinitiate if there is no improvement for a certain number of consecutive non-improving iterations. Improved solutions are accepted, as well as worse solutions based on the current level of the rGD-cNbs. Furthermore, a composite neighbourhood structure is employed within the rGD-cNbs in order to help the algorithm to better explore the search space. This study involves the evaluation of approaches on 18 benchmark datasets that are available in UCI machine learning repository. Experimental results show that the rGD-cNbs is able to achieve competitive results in comparison with other available meta-heuristic approaches in the literature in terms of the minimal reduct.
关键词：Artificial intelligence; attribute reduction;rough set theory;great deluge;composite neighbourhood structures
