Zhang Qiang, Fan Jiahui, Gao Chaobang
School of Computer Science, Chengdu University, Chengdu, 610106, China.
Key Laboratory of Pattern Recognition and Intelligent Information Processing, Institutions of Higher Education of Sichuan Province, Chengdu University, Chengdu, China.
Sci Rep. 2024 Oct 23;14(1):25094. doi: 10.1038/s41598-024-75992-z.
In multi-criteria decision-making and model evaluation, determining the weight of criteria is crucial. With the rapid development of information technology and the advent of the big data era, the need for complex problem analysis and decision-making has intensified. Traditional CRiteria Importance Through Intercriteria Correlation (CRITIC) methods rely on Pearson correlation, which may not adequately address nonlinearity in some scenarios. This study aims to refine the CRITIC method to better accommodate nonlinear relationships and enhance its robustness. We have developed a novel method named CRiteria Importance Through Intercriteria Dependence (CRITID), which leverages cutting-edge independence testing methods such as distance correlation among others. This approach enhances the assessment of intercriteria relationships. Upon application across diverse data distributions, the CRITID method has demonstrated enhanced rationality and robustness relative to the traditional CRITIC method. These improvements significantly benefit multi-criteria decision-making and model evaluation, providing a more accurate and dependable framework for analyzing complex datasets.
在多标准决策和模型评估中，确定标准的权重至关重要。随着信息技术的快速发展和大数据时代的到来，对复杂问题分析和决策的需求日益增强。传统的基于标准间相关性的标准重要性（CRITIC）方法依赖于皮尔逊相关性，在某些情况下可能无法充分处理非线性问题。本研究旨在改进CRITIC方法，以更好地适应非线性关系并增强其稳健性。我们开发了一种名为基于标准间依赖性的标准重要性（CRITID）的新方法，该方法利用了诸如距离相关性等前沿独立性测试方法。这种方法增强了对标准间关系的评估。在跨不同数据分布应用时，CRITID方法相对于传统CRITIC方法表现出更高的合理性和稳健性。这些改进极大地有利于多标准决策和模型评估，为分析复杂数据集提供了一个更准确、可靠的框架。
