Hesitant fuzzy linguistic term sets (HFLTSs) are applied to deal with situations in which people are hesitant in providing linguistic evaluations and have been widely used in qualitative group decision making processes. Considering the fact that the correlation coefficient has been widely used in many research domains, in this paper, we first present a novel correlation coefficient for HFLTSs. The significant features of the proposed correlation coefficient are that the HFLTSs need not to be extended to the same length and the correlation coefficient lies in \([-1,1]\). Two examples are used to illustrate the effectiveness of the proposed correlation coefficient. Secondly, based on the proposed correlation coefficient and the collaborative filtering algorithm, we develop a new method to deal with incomplete hesitant fuzzy linguistic information. Furthermore, by analysing the correlation coefficient matrix, we present a new method for determining experts’ weights based on the concept of ‘attenuation’ and ‘how much trust’ they obtained. At last, a case study is used to evaluate the performance of our method.

犹豫的模糊语言术语集（HFLTS）用于处理人们犹豫不决地提供语言评估的情况，并已广泛用于定性的群体决策过程中。考虑到相关系数已在许多研究领域中得到广泛应用的事实，在本文中，我们首先提出一种新的HFLTS相关系数。所提出的相关系数的显着特征是HFLTS不需要扩展到相同的长度，并且相关系数位于\（[-1,1] \）。使用两个例子来说明所提出的相关系数的有效性。其次，在提出的相关系数和协同过滤算法的基础上，提出了一种处理不完全犹豫模糊语言信息的新方法。此外，通过分析相关系数矩阵，我们提出了一种基于“衰减”和“获得多少信任”的概念确定专家权重的新方法。最后，通过案例研究来评估我们方法的性能。