Abstract: 【Purpose/significance】Aiming at the practical problems of discretization, fuzziness and complexity of network rumor refuta⁃ tion information dissemination. This article establishes a theoretical framework for network anti rumor information profiling and gover⁃ nance, which includes "data layer - portrait layer - pattern layer". It aims to provide theoretical guidance for the practice of network rumor refutation and information governance【. Method/process】Based on the tipping point theory, this article analyzes the user profile, content profile, and environmental profile of online rumor refutation information, starting from three major principles: individual char⁃ acter rule, adhesion factor rule, and environmental power rule. It explores the impact of different combinations of factors on the gover⁃ nance mechanism of online anti rumor information【. Result/conclusion】This study found that the public has a deep understanding of the theme of online anti rumor information, and positive emotions dominate. A single factor does not constitute a necessary condition for the outbreak of online rumor refutation information. The causal path that triggers online rumor refutation information is multiple and concurrent. This article summarizes three types of network anti rumor information governance models: user assistance model, con⁃ tent oriented model, and environment driven model.【Innovation/limitation】This article establishes a theoretical framework for the por⁃ trayal and governance of online anti rumor information. It implements a research approach from discretizing data to conceptualizing patterns. The limitation of this article is that it does not subdivide the types of online rumor refutation events and does not consider the impact of time factors on the effectiveness of online rumor refutation information governance.