Summary: Based on probabilistic uncertain linguistic term set, a group decision-making approach is proposed considering the behavior of decision makers, which is used for dealing with uncertain linguistic multi-attribute group decision-making problem with completely unknown attribute weights. First, the hamming distance between any two probabilistic uncertain linguistic term sets is defined. A maximum deviation model is established based on the distance between each alternative and the negative ideal solution. All the attribute weights are obtained by solving the model. Second, the classical TODIM (an acronym in Portuguese for interactive and multi-criteria decision-making) method is extended and improved, and the rank of all the initiative is gotten by calculating the prospect value of each initiative based on the established TODIM method. Finally, a selection problem of the initiatives is presented to illustrate the practicality of the proposed method. Analyzing the sensitivity of the parameter and comparing it with the TOPSIS (the technique for order of preference by similarity to ideal solution) method, it is revealed to be more effective.