In the Internet era, accurately establishing the propagation model of social network public opinion is of great help to the guidance and control of public opinion. Based on the social network information dissemination model and node identification algorithm, this paper organically combines the ant colony algorithm and simulated annealing algorithm, constructs the optimal path selection model for social network information dissemination, and designs a set of DAILY models for public crisis information dissemination adapted to the Internet information environment and technology, so as to realize the efficiency improvement of data flow. The accuracy of the model in this paper has obvious advantages compared with the traditional SIR model, and the model fitting curve is basically consistent with the real data curve, and its absolute error value and RMSE value are low. Meanwhile, the simulation results show that the proposed important node recognition algorithm has high accuracy and feasibility. Compared with the classical ant colony algorithm and Dijkstra’s algorithm, the model in this paper can disseminate a larger amount of information per unit time and has a higher accuracy rate, with the highest accuracy rate of 90.32% in the three groups of experiments, which is able to find the optimal path of information dissemination, and has an important guiding role in analyzing and guiding the corresponding public opinion.
