越览（205）——精读期刊论文的4. 结论

分享兴趣，传播快乐，

增长见闻，留下美好。

亲爱的您，这里是LearingYard学苑！

今天小编为您带来“越览（205）——精读期刊论文

《基于累积前景理论的多粒度概率语言双边匹配决策方法》的

4. 结论。

欢迎您的访问！

Share interest, spread happiness,

increase knowledge, and leave beautiful.

Dear, this is the LearingYard Academy!

Today, the editor brings the

＂Yue Lan (205)：Intensive reading of the journal article

＇A multi-granularity probabilistic language bilateral

matching decision-making method

based on cumulative prospect theory’

4. Conclusion.

Welcome to visit!

一、内容摘要（Summary of Content）

本期推文将从思维导图、精读内容、知识补充三个方面介绍精读期刊论文《基于累积前景理论的多粒度概率语言双边匹配决策方法》的4. 结论。

This issue of tweets will introduce 4. Conclusion of “A multi-granularity probabilistic language bilateral matching decision-making method based on cumulative prospect theory ” from three aspects: mind map, intensive reading content, and knowledge supplement.

二、思维导图（Mind map）

三、精读内容（Intensive reading content）

第一，总结了本文的研究结论。本文表明 MPLANC 能够对多粒度概率语言信息进行简单而有效的统一，相较于现有统一粒度方法，能够在转换过程中避免原始信息的丢失，从而为后续决策分析提供更加可靠的信息基础。

First, the research findings of this paper are summarized. This paper demonstrates that MPLANC can achieve a simple and effective unification of multi-granularity probabilistic linguistic information. Compared with existing unification granularity methods, it can avoid the loss of original information during the transformation process, thus providing a more reliable information foundation for subsequent decision analysis.

其次，提出本文的研究内容。针对多粒度概率语言信息环境下属性权重未知的问题，本文构建了基于 MPLANC 双向投影与最大熵原理的非线性优化模型，实现了属性权重的客观计算；同时提出了 MPLANC 幂 HM 算子，使得由此获得的正、负理想点更加合理，增强了决策结果的区分性与稳定性。

First, the research findings of this paper are summarized. This paper demonstrates that MPLANC can achieve a simple and effective unification of multi-granularity probabilistic linguistic information. Compared with existing unification granularity methods, it can avoid the loss of original information during the transformation process, thus providing a more reliable information foundation for subsequent decision analysis.

接着，在上述研究基础上，指出本文创新点。本文进一步提出了思考匹配主体不同心理偏好的 MPLANC 累积前景理论双边匹配方法。通过服务外包平台问题的算例分析、灵敏度分析和比较分析，验证了所提出方法在实际应用中的实用性与灵活性。研究结果表明，该方法能够有效反映不同风险程度下匹配主体的心理机制，使得匹配结果更加符合主体真实意愿，例如在企业产品供需匹配等高风险情境下，模型能够体现主体“避险强于趋利”的决策特征。

Next, based on the above research, the innovative points of this paper are pointed out. This paper further proposes a bilateral matching method based on MPLANC cumulative prospect theory, considering the different psychological preferences of the matching subjects. Through case studies, sensitivity analyses, and comparative analyses of the service outsourcing platform problem, the practicality and flexibility of the proposed method in real-world applications are verified. The results show that this method can effectively reflect the psychological mechanisms of the matching subjects under different risk levels, making the matching results more consistent with the subjects' true intentions. For example, in high-risk scenarios such as matching supply and demand for enterprise products, the model can reflect the subject's decision-making characteristic of “risk aversion being stronger than profit seeking.”

最后，思考了未来研究方向。本文为实际双边匹配问题提供了一种新的分析思路，但仍存在必定局限性，即仅思考了主体在单一阶段下的偏好特征。鉴于主体偏好在不同阶段可能发生变化，未来研究将进一步拓展至多阶段双边匹配问题，以期在动态决策环境下实现主体满意度的最大化。

Finally, future research directions were considered. This paper provides a new analytical approach to practical bilateral matching problems, but it still has certain limitations, namely, it only considers the preference characteristics of subjects in a single stage. Given that subject preferences may change at different stages, future research will be further extended to multi-stage bilateral matching problems, aiming to maximize subject satisfaction in a dynamic decision-making environment.

四、知识补充（Knowledge supplement）

为便于理解本文所提出方法的研究背景与建模思路，现从多粒度概率语言信息、属性权重确定以及决策主体行为特征等方面，对相关知识进行补充说明。

To facilitate understanding of the research background and modeling approach of the proposed method, we will now provide supplementary explanations on relevant knowledge from aspects such as multi-granularity probabilistic linguistic information, attribute weight determination, and behavioral characteristics of decision-making agents.

1. 多粒度概率语言信息的表达特点（Characteristics of multi-granular probabilistic language information representation）

在多属性群决策与双边匹配问题中，不同决策主体由于知识背景、经验水平和表达习惯的差异，往往采用不同粒度的概率语言术语对方案进行评价。这种多粒度概率语言信息能够更灵活地刻画评价不确定性，但也导致不同主体评价结果之间难以直接比较。因此，有必要通过统一处理机制将多粒度概率语言信息转换到同一评价框架下，以保证后续分析的可行性和一致性。

In multi-attribute group decision-making and bilateral matching problems, different decision-makers often use probabilistic language terms of varying granularities to evaluate solutions due to differences in knowledge background, experience level, and expression habits. While this multi-granularity probabilistic language information can more flexibly characterize evaluation uncertainty, it also makes direct comparison of evaluation results between different entities difficult. Therefore, it is necessary to transform this multi-granularity probabilistic language information into a unified evaluation framework through a standardized processing mechanism to ensure the feasibility and consistency of subsequent analyses.

2. 多粒度信息统一处理的必要性（The necessity of unified processing of multi-granularity information）

如果忽略语言粒度差异，直接对评价信息进行聚合，容易引入偏差甚至导致信息失真。合理的统一方法应在保证语义一致性的前提下，尽量保留原始评价信息。MPLANC 方法通过对概率语言信息的统一转换，为多粒度评价信息的综合处理提供了基础条件，有助于提升决策结果的可靠性。

Ignoring differences in linguistic granularity and directly aggregating evaluation information can easily introduce biases and even lead to information distortion. A reasonable unification method should preserve as much of the original evaluation information as possible while ensuring semantic consistency. The MPLANC method, through the unified transformation of probabilistic linguistic information, provides the foundation for the comprehensive processing of multi-granularity evaluation information, which helps improve the reliability of decision-making results.

3. 属性权重未知情况下的客观赋权问题（Objective weighting problem when attribute weights are unknown）

在实际决策中，属性权重往往难以通过主观方式准确给出，尤其是在评价信息本身存在不确定性的情况下。采用客观赋权方法，可以减少人为因素对决策结果的影响，并使权重结果更加稳定。通过构建合理的权重计算模型，能够在充分利用现有评价信息的基础上，得到具有解释性的属性权重，为后续方案排序与匹配分析提供支持。

In practical decision-making, attribute weights are often difficult to accurately determine subjectively, especially when the evaluation information itself is uncertain. Adopting objective weighting methods can reduce the influence of human factors on decision outcomes and make the weight results more stable. By constructing a reasonable weight calculation model, interpretable attribute weights can be obtained by fully utilizing existing evaluation information, providing support for subsequent scheme ranking and matching analysis.

4. 属性相关性在决策分析中的作用（The role of attribute correlation in decision analysis）

在多属性决策问题中，各属性之间一般并非完全独立，而是存在必定程度的相关性或交互影响。如果忽略属性相关性，可能会高估或低估部分属性的作用，从而影响决策结果的合理性。将属性相关性纳入决策模型，有助于更全面地反映复杂决策环境的实际特征。

In multi-attribute decision-making problems, the attributes are usually not completely independent, but rather have a certain degree of correlation or interaction. Ignoring attribute correlation may overestimate or underestimate the role of some attributes, thus affecting the rationality of the decision outcome. Incorporating attribute correlation into the decision-making model helps to more comprehensively reflect the actual characteristics of complex decision-making environments.

5. 决策主体风险偏好的影响（The impact of decision-making entities' risk preferences）

大量研究表明，决策主体在面对不同风险水平的决策情境时，往往表现出非对称的风险态度。在高风险情境下，主体一般更倾向于规避损失，而在低风险情境下则可能表现出必定的风险偏好。引入能够刻画风险态度的行为决策思想，有助于更真实地反映主体的心理机制，使匹配结果更加符合实际决策行为。

Numerous studies have shown that decision-makers often exhibit asymmetric risk attitudes when faced with decision-making situations of varying risk levels. In high-risk situations, individuals typically tend to avoid losses, while in low-risk situations they may show a certain degree of risk preference. Introducing behavioral decision-making concepts that can characterize risk attitudes helps to more accurately reflect the psychological mechanisms of individuals, making the matching results more consistent with actual decision-making behavior.

今天的分享就到这里了。

如果您对文章有独特的想法，

欢迎给我们留言，让我们相约明天。

祝您今天过得开心快乐！

That's all for today's sharing.

If you have a unique idea about the article,

please leave us a message,

and let us meet tomorrow.

I wish you a nice day!

翻译：谷歌翻译

参考资料：百度百科、Chat GPT

参考文献：王磊,李文杰,王海.基于累积前景理论的多粒度概率语言双边匹配决策方法[J].控制与决策, 2025, 40(1): 300-307.

本文由LearningYard学苑整理发出，如有侵权请在后台留言！

文案|yyz

排版|yyz

审核|hzy