Research on the "multi-agent co-governance" system of unfair competition on internet platforms: Based on the perspective of evolutionary game.

Unfair competition on internet platforms (UCIP) has become a critical issue restricting the platform economy's healthy development. This paper applies evolutionary game theory to study how to utilize multiple subjects' synergy to supervise UCIP effectively. First, the "multi-agent co-governance" mode of UCIP is constructed based on the traditional "unitary supervision" mode. Second, the government and internet platform evolutionary game models are built under two supervision modes. Finally, MATLAB is used to simulate and analyze the evolutionary stage and parameter sensitivity. In addition, we match the model's evolutionary stage with China's supervisory process. The results show that (1) the Chinese government's supervision of UCIP is in the transitional stage from "campaign-style" to "normalization." (2) Moderate government supervision intensity is essential to guide the game system to evolve toward the ideal state. If the supervision intensity is too high, it will inhibit the enthusiasm for supervision. If the supervision intensity is too low, it cannot form an effective deterrent to the internet platforms. (3) When the participation of industry associations and platform users is low, it can only slow down the evolutionary speed of the game system's convergence to the unfavorable state. Nevertheless, it cannot reverse the evolutionary result. (4) Maintaining the participation level of industry associations and platform users above a specific threshold value while increasing punishment intensity will promote the transition of government supervision from the "campaign-style" to the "normalization" stage. This paper provides ideas and references for the Chinese government to design a supervision mechanism for UCIP. [ABSTRACT FROM AUTHOR]