Abstract: Multi criteria decision making with intuitionistic fuzzy sets is a hot topic in the field decision making in recent years. Intuitionistic fuzzy sets use membership degree, non-membership degree and hesitation degree to represent fuzzy decision information, which can more flexibly represent and process fuzzy information. Because of the difficulty of determining the three critical degrees, Atanassov further extends the concept of intuitionistic fuzzy set and proposes interval valued intuitionistic fuzzy sets, which has been widely used in logistics management, energy analysis, decision-making analysis and many other fields. The score function of interval intuitionistic fuzzy sets is an important way to transform interval fuzzy values into accurate values. However, the score function of interval intuitionistic fuzzy sets also has the problems of anti-intuitionistic scoring results and ignoring interval information, which affect the reliability and objectivity of group preference aggregation. Therefore, this paper first proposes an improved score function based on set pair theory on the basis of analyzing the defects of the existing scoring function of interval intuitionistic fuzzy sets. Secondly, based on the improved scoring function of interval intuitionistic fuzzy sets, this paper designs an interval intuitionistic fuzzy dynamic group decision-making model based on the scoring function, and applies it to patent quality evaluation, providing a good information fusion framework for patent quality evaluation. From the perspective of application, patent quality evaluation involves many disciplines, with strong cross cutting in knowledge fields, and the trend of diversified evaluation indicators is increasingly obvious. Therefore, patent quality evaluation is usually a multi-attribute group decision-making problem involving multiple experts. Through the analysis of the existing evaluation index system, this paper intends to evaluate the patent quality from three aspects: patent technology level, patent transformation benefits, and patent protection regulations.
　　In this paper, based on the analysis of the shortcomings of the existing score function of interval intuitionistic fuzzy sets, an improved score function based on set pair theory is proposed; Secondly, based on the improved score function of interval intuitionistic fuzzy sets, this paper designs an interval intuitionistic fuzzy dynamic group decision-making model based on the scoring function, and applies it to patent quality evaluation, providing an effective information fusion framework for patent quality evaluation. This paper intends to evaluate patent quality from three aspects: patent technology level, patent transformation benefits, and patent protection regulations. Among them, (1)Technology level is the core indicator of patent quality, which can reflect the endogenous quality of a patent. According to the Patent Law, a patent can only be granted if the technology has novelty, creativity and practicability. (2)The benefit of patent transformation refers to the sum of the economic value generated by the application of a patented technology in the market process. It is the embodiment of the market value of the patented technology and the ultimate destination of the patented technology: it is transformed into productivity, generates economic benefits and contributes value to the society. (3)Patent protection regulations are the basic conditions to ensure the quality of patents and provide strong legal protection for innovation subjects or patent holders.
This paper summarizes the existing score functions of interval intuitionistic fuzzy sets, and analyzes the shortcomings of the existing score functions from two aspects: whether the evaluation value can follow the decision-maker's intuition and whether the evaluation value can reflect the interval fuzzy information. In order to improve the scoring and ranking ability of score function for interval intuitionistic fuzzy sets, based on the concept of set pair potential, this paper proposes an improved score function for interval intuitionistic fuzzy sets, and proves that it satisfies the basic intuition of decision makers. Driven by group consensus, this paper designs an interval intuitionistic fuzzy information fusion model based on scoring function, and applies it to the patent quality evaluation model, and achieves good evaluation results.

Key words: interval-valued intuitionistic fuzzy set, score function, dynamic group decision making, patent quality

摘要： 记分函数是对区间直觉模糊集进行比较和集结的重要工具,其排序能力直接影响评价结果的优劣。本文在总结整理现行的区间直觉模糊集记分函数的基础上,从直觉符合性和能否描述决策者模糊判断两个层面,系统分析了现有记分函数的不足。并在此基础上,引入集对分析方法构建改进的区间直觉模糊集记分函数,并证明了其相关性质。根据提出的记分函数,本文设计区间直觉模糊型动态群决策方法,并将其应用于专利质量评价中。

关键词: 区间直觉模糊集, 记分函数, 动态群决策, 专利质量