LAUSR

Abstract Public opinions play a pivotal role in public participation and decision-making in mega construction projects. The rise of technology and social media has facilitated the online public debate on social impacts derived from mega construction projects. This study presents a novel approach to analyze the evolution of public opinions through social media. This analysis framework covers: the temporal patterns of online messages, evolution of frames, and evolution of users. Such novel approach is employed to analyze the evolution of online public opinions on a mega construction project. Content analysis and a number of curve fitting methods were innovatively combined. Results show that social event had dual diffusion cycles while the formation mechanism of each cycle was completely different. The key parameters of Bimodal Gaussian model can depict the characteristics of online diffusion of public opinions. Elites and media organizations form the crucial nodes in the online diffusion process. The sequence of each frame and their share in the total number of frames are determined by preference of elites and media organizations. These findings are useful to develop effective strategy for public engagement, and to prevent and intervene the social conflicts.