Intuitionistic fuzzy linguistic clustering algorithm based on a new correlation coefficient for intuitionistic fuzzy linguistic information

For observations to be classified, when scoring rules are imprecise or the cost of their computation is too high, the clustering method under linguistic information is necessary. Considering the accuracy of intuitionistic fuzzy linguistic variable in expressing experts’ opinions, a clustering algorithm is presented in this paper. Firstly, the concept of triangular intuitionistic fuzzy linguistic variables (TIFLVs) is introduced, and a new formula is developed for calculating correlation coefficient of TIFLVs. Then, the correlation coefficient plays a central role in our modified λ-cutting algorithm for clustering, which is utilized to construct an equivalence correlation matrix. In addition, a silhouette cluster validity index of TIFLVs is proposed to revise the results of clustering. Finally, the experimental results demonstrate the application and practicability of the clustering algorithm.