Abstract: 【 Purpose/significance】 Aiming at the current situation of high demand for knowledge of users health information, various in⁃ formation contents, lack of rationality of organization and narrow application scope in online health community and online diagnosis and treatment platform, on the basis of constructing a knowledge graph of gynecological diseases that is suitable for resource character⁃ istics and user demand characteristics, a faceted retrieval system model is designed to help women provide independent choices and demand discovery pathways in private disease search. 【Method/process】 Through the research on knowledge needs of gynecological diseases, the top-level construction framework of the knowledge graph of gynecological diseases is proposed, based on the analysis of the entity types and relationship types among entities in the knowledge graph, the knowledge graph for gynecological diseases is con⁃ structed according to the steps of knowledge extraction, knowledge fusion, knowledge storage and visualization, and a faceted system is constructed, which combines the definition of disease condition, diagnosis and treatment, self-management and other categories, trying to dynamically display the faceted results in combination with the needs of users.【 Result/conclusion】 Based on the situation of pelvic inflammatory disease knowledge graph, taking 39. net as an example to carry out faceted system, and prototype demonstrations and comparative analysis were conducted to verify the feasibility and effectiveness of the model, providing reference for improving the exist⁃ ing faceted system for online health communities.【 Innovation/limitation】 Lack of comparative analysis in large-scale and multi sce⁃ narios.