| This thesis researches on the probabilistic hesitant fuzzy multiple attribute decision making method based on bipartite projection network.Due to the ability of bipartite network to express complex data structure,bipartite network is applied to probabilistic hesitant fuzzy sets(PHFSs)theory.It not only enriches the probabilistic hesitant fuzzy multi-attribute decision-making method,but also extends the application of bipartite network.The combination of the two methods can preserve the decision information and improve decision quality to get a more scientific and reasonable decision result.This paper mainly focuses on the following researches.(1)A probabilistic hesitant fuzzy multiple attribute decision making problem with completely unknown probability information is studied.Aiming at the case that the probabilities of probabilistic hesitant fuzzy elements(PHFEs)are completely unknown,a nonlinear programming model is proposed to determine the unknown probabilities of PHFEs.According to the evaluation of decision-makers,the weights of attributes can be obtained by entropy weight method.In addition,an approach based on LINMAP method is put forward to determine the probabilistic hesitant fuzzy ideal solutions objectively under probabilistic hesitant fuzzy environment.Then,the compromise ratio method(CRM)is utilized in ranking the alternatives.(2)A probabilistic hesitant fuzzy recommendation decision-making method based on bipartite network projection is proposed,which considers the users’ decisionmaking information and experts’ evaluation values.Firstly,according to users’ decision-making information and the experts’ evaluation values,a bipartite graph connecting users and alternatives is established.Secondly,the satisfaction degree of PHFEs is proposed to calculate the satisfaction degree of alternatives.The satisfaction degree of the alternatives is treated as the edge weight of bipartite network projection,and the recommended alternative is determined according to the distribution of the final projection resource.(3)A clustering method based on bipartite network projections and PHFSs is proposed.Firstly,a novel method is given to calculate the correlation degree for different alternatives by investigating the relationship between nodes in the same set.The edge weight of bipartite graph is depicted by a PHFE.And the correlation degrees for alternatives derived by our method are still PHFEs.Then,to measure the information capacity of probabilistic hesitant fuzzy element,a novel concept called probabilistic hesitant fuzzy information energy is proposed.Finally,based on the obtained information energy matrix,a netting method is adopted to cluster the alternatives.(4)A new multi-stage probabilistic hesitant fuzzy multiple attribute decision making method based on interactive bipartite network is proposed.Firstly,an interactive bipartite graph projection network is established by considering the objective attributes of alternatives and the subjective demand of users.And a model based on the satisfaction degree of alternatives is established to compute the weights of the initial resource values in different stages.The adjustment suggestion for alternatives can be obtained from the changes of attribute weights. |
