1. 融合差分教学优化的粗糙集属性约简算法.
- Author
-
周婉婷, 郑颖春, and 魏博涛
- Abstract
To address the challenges of high computational complexity and the tendency to get stuck in local optima during at- tribute reduction within traditional rough set theory, this paper proposed an innovative rough set attribute reduction algorithm based on differential teaching-learning optimization (AR-DTLBO). Leveraging the global search capabilities of the differential teaching-learning optimization algorithm along with the strengths of rough set theory in handling imprecise and uncertain data, the algorithm aimed to optimize the process. Firstly, it enhanced the teaching-learning optimization algorithm by introducing an adaptive teaching factor and a differential mutation strategy, thereby enhancing its search capabilities and optimization performance. Subsequently, it refined the attribute reduction process through the improved teaching-learning optimization algorithm s teaching and learning phases, effectively reducing the dimensionality and complexity of datasets. Finally, it conducted comparative experiments between the proposed AR-DTLBO algorithm and six other algorithms, using eight datasets from the UCI database. The experimental results demonstrate that the proposed algorithm achieves favorable outcomes in terms of reduction length, reduction time, reduction rate, and classification accuracy. This successful reduction and optimization of datasets not only reduces redundant information but also enhances the precision of decision rules. These findings provide valuable sup- port for decision analysis, data mining, and other related fields. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF