Research on Military Knowledge Graph Fusion Method Based on Multi-Technology

In the context of information warfare, the fusion of knowledge graphs in the military field is of great significance for assisting decision-making, intelligence analysis, and battlefield situational awareness. This paper proposes a method for constructing and applying military knowledge graphs based on multi-technology fusion, addressing the fusion issues between the CMNEE dataset and external knowledge graphs (such as CN-DBpedia). The method encompasses four key steps: data preprocessing, entity linking, graph construction, and graph fusion. By integrating technologies such as Stanford CoreNLP, TF-IDF with cosine similarity, Graph Convolutional Networks (GCN), BiLSTM + CRF models, and knowledge embedding (TransE), the fusion effect is significantly enhanced. Experimental results show that this method outperforms other combination methods across multiple metrics, providing strong support for military knowledge mining and application, with important practical significance and broad application prospects.