Abstract: 【Purpose/significance】To organize and mine information and knowledge in the field of modern revolutionary wars, this study aims to construct an event-centric knowledge graph for revolutionary war events and explore its innovative applications in public library resource development and services.【Method/process】Taking the Menglianggu Campaign as a case study, entity and relation⁃ ship extraction was conducted using the ChatGLM-6B model to build a revolutionary wars knowledge graph. The research further in⁃ vestigated innovative application pathways for this knowledge graph in public library services.【Result/conclusion】The study proposes an event-centric methodology for constructing revolutionary wars knowledge graphs, achieving intelligent organization and associative mining of red culture resources. It explores the application value and pathways of the knowledge graph in public libraries, including: resource digitization and integration, intelligent retrieval and navigation for resource organization, red culture education, culturaltourism integration services, and red-themed research activities.【Innovation/limitation】The knowledge extraction method based on large language models demonstrates promising performance in low-resource settings. While the proposed event-centric methodology advances knowledge graph construction, it requires significant computational resources and relies solely on a single large language model, lacking comparative analysis with alternative models.