报告题目：Building Interpretable Fuzzy Sets Using the Deck-of-Cards Method

报告摘要：This talk presents a socio-technical and co-constructive framework designed to actively involve decision makers in the construction of fuzzy membership functions (MFs). Traditional approaches to defining MFs—based either on expert opinion or data-driven methods—often suffer from subjectivity, lack of interpretability, and limited interaction between analysts and decision makers. The proposed DoC-MF approach adapts the Deck-of-Cards (DoC) method, to the context of fuzzy set modeling. It enables the collaborative elicitation of scales of values and shapes of membership functions that reflect human perception more faithfully.

The methodology unfolds in three steps: (1) constructing a scale of values using blank cards to represent perceived intensity differences among performance levels; (2) determining the core (full confidence interval) and support (non-zero membership region) through a co-constructive threshold protocol; and (3) shaping the left and right sides of the MF using mini DoC procedures that capture how membership increases or decreases. The resulting MFs are interpretable, traceable to the elicitation process, and tailored to each decision maker’s semantics.

The talk highlights how DoC-MF improves interpretability, reduces bias and inconsistency, and aligns fuzzy models with the real cognitive structures of human reasoning.

报告人简介：Luis Martinez Lopez 教授是西班牙哈恩大学计算机科学系教授, 主要研究方向为模糊决策、词计算、多准则决策等，目前担任 International Journal of Computational Intelligence Systems 主编，Information Sciences、Expert System with Applications、Knowledge-based System、IEEE Transactions on fuzzy systems、Information Fusion 等期刊副主编，IFSA 会士（IFSA Fellow），IEEE senior member，欧洲模糊逻辑与技术学会（ESFLT）会士。发表 270 余篇文章，2017-2024 年连续入选全球高被引学者榜单。