HNLOWA集成算法应用于推荐系统评估

针对复杂决策环境下的决策者倾向于使用定性评价数据和现实环境中数据信息通常存在正态分布规律的问题，构建了基于犹豫正态语言有序加权平均（hesitant normal linguistic ordered weighted average，HNLOWA）算子的决策模型。将正态模糊数和犹豫模糊语言元相结合，引入了犹豫正态语言元（hesitant normal linguistic element，HNLE）的概念，其不仅能够运用语言变量来描述决策信息，还能传递出语言决策信息的分布情况；定义了HNLE之间的基本运算法则和大小判别准则，并设计HNLOWA算子用于对HNLE进行信息融合，同时探究了HNLOWA算子的相关性质；构建了基于HNLOWA集成算子的多属性决策方法，并运用推荐系统选择评估案例进行了模型的验证分析。

HNLOWA Aggregation Algorithm and Its Application to Recommendation System Evaluation

In view of the fact that decision-makers tend to use qualitative evaluation data in complex decision-making environment and the normal distribution of data information in real environment, a decision-making model based on hesitant normal linguistic ordered weighted average（HNLOWA） operator is constructed. Firstly, the concept of hesitant normal linguistic element（HNLE） is introduced by combining normal fuzzy values with hesitant fuzzy linguistic element, which can not only use linguistic variables to describe decision-making information, but also transfer the distribution of linguistic decision-making information. Secondly, the basic operation rules and ranking method between HNLEs are defined, and the HNLOWA operator is designed to fuse HNLE information. In addition, the related properties of HNLOWA operator are explored. Finally, a multi-attribute decision-making method based on HNLOWA operator is developed, and an example about the recommendation system selection evaluation is provided to verify the proposed model’s practicability.