Abstract: 【Purpose/significance】The core users identification of online health community is important for the sustainable develop⁃ ment of the community and the dissemination of scientific health information in the society.【Method/process】The core users identifi⁃ cation method proposed in this paper integrates user attribute features and social network features. Firstly, we use the Node2vec net⁃ work representation learning model to map the nodes in the user's social network as vectors, and select the optimal clustering algo⁃ rithm to cluster and filter the user node vectors, and finally introduce the customer value segmentation model (RFM model) in the field of marketing to construct the core user influence evaluation model integrating the social network characteristics, and then we obtain the core user ranking of the online health community after the comprehensive analysis of the candidate set.【Result/conclusion】We selected the data in DingXiangYuan Forum to validate the method proposed in this paper, and proved the effectiveness of the method proposed in this paper, and the results show that the method can better identify the core users of online health community.【Innovation/ limitation】We introduced the RFM model to consider the influence of multi-dimensional key indicators such as time, frequency and value factors, which can better characterize and identify core users from the perspective of user value, but does not adequately consider the dynamic evolution process of the influence of core users, so we can consider combining with the time dimension to further explore the dynamic mechanism of core user identification in the future