A large-scale group decision making method with text mining and probabilistic linguistic complementation for energy transition path assessment

Selecting a scientific, practical and efficient energy transition path is the key to solving the main contradiction in the energy industry. Considering text mining and probabilistic linguistic information complementation, a large-scale group decision-making method is developed. Firstly, text mining technology is used to extract big data of public behavioral preference, so as to establish the evaluation criteria system of energy transition paths, and a criterion weighting model is proposed according to affinity coefficient and TextRank algorithm. Then, experts are clustered based on the social trust network analysis, so that the missing probabilistic linguistic information of expert are completed. Next, the clusters with overlapping features or isolated nodes are optimized via the principle of minimum deviation, and the expert weights are modified by combining information similarity. Finally, the alternatives are ranked by the S-hyperbolic absolute risk aversion utility function. The proposed method is applied to the practical problem of evaluating China's energy transition paths under the dual-carbon goal, with “enhancing the use of clean energy” as the optimal path. The validity and practicability of the model is demonstrated through the multidimensional sensitivity analysis, and insights and suggestions are given in this field.