An enhanced SIR dynamic model: the timing and changes in public opinion in the process of information diffusion

Social media platforms provide great convenience for public share and get access to various information, which bring challenges to both companies and company to manage negative impacts of public opinion. When an event happens, truth and rumors are intertwined in the process. To figure out how rumor influences the information diffusion process timely and identify points when necessary actions should be taken to control the impacts of rumors, an enhanced S–I–R (susceptible–infectious–recovered) dynamic model, involving rumors occurring sequentially in information diffusion process, is proposed. Based on the proposed model, we develop dynamic equations with the spreading probability and weight of the intervening acts. Then, we simulate how the process works theoretically where simulation in two different group of real-world datasets is conducted to demonstrate the validity of the proposed model. Besides, the present study illustrates that different spreading rates and weights of intervening acts could result in different diffusion situations, especially when the weight induces an increase in the peak point (ω 2 = 0.5) and secondary climax(ω 4 = 0.6). Finally, the present study provides suggestions practically for company in terms of managing rumor in online public opinion from different aspects. As our paper extend the research process of online public opinion to take rumor into consideration at two time periods, it not only enriches models of online public opinion diffusion process, but also shed lights on further studies concerning on online public opinion diffusion.