Abstract: 【Purpose/significance】In the era of convergence media, information dissemination is characterized by multi-form, multichannel and multi-interaction. The responsive information dissemination network is becoming more and more complicated, which brings great challenges to public opinion information control and traceability. Efficient and accurate traceability algorithm has become a technical problem to be solved urgently.【Method/process】Aiming at the problem of information traceability of convergence media in the era of big data, a two-stage information traceability algorithm was proposed based on the construction of the information dissemi⁃ nation model of convergence media and combined with the center of mass theory. The central index and centroid algorithm were used to find the key nodes that make public opinion events break out, and the computing power and stability analysis of the proposed algo⁃ rithm were carried out through simulation verification.【Result/conclusion】The results show that compared with other different types of algorithms, the proposed traceability algorithm can adapt well to the communication characteristics of the convergence media plat⁃ form, and can not only respond quickly, but also maintain a high accuracy rate.【Innovation/limitation】The center of mass theory is in⁃ novatively applied to the field of public opinion information traceability, and a new traceability algorithm suitable for the convergence media platform is built. However, the actual information transmission model of the convergence media platform is more complex, and empirical research is still needed to be carried out for iterative improvement.