School of Business, Central South University
A new method of preference information fusion is proposed to solve the problem of attribute correlation in large group emergency decision making. Firstly, the optimal discrete fitting model is used to measure the risk preference of experts, and proposes a hesitant fuzzy meta-supplement method considering the risk preference of experts. Secondly, TF-IDF algorithm is used to obtain the interrelated event attribute set. Then, combining the traditional principal component analysis method and error theory, a principal component analysis model (PCA) based on hesitating fuzzy language is proposed to obtain several unrelated principal attributes and their weights, and then information aggregation and alternatives selection are carried out. Finally, the feasibility and effectiveness of the proposed method are verified by the flood disaster in Guangxi.