As a precious cultural heritage of Chinese civilization, the study of oracle bone inscriptions is of great significance to the fields of philology, history and archaeology. This paper focuses on the construction of oracle bone inscription knowledge graph and its application in historical language analysis, focusing on the text organization, grammatical structure analysis and semantic information mining of oracle bone inscriptions. By adopting natural language processing and machine learning methods, this paper realizes the automatic annotation and deep semantic relationship mining of oracle bone inscriptions. At the same time, the construction of cross-modal knowledge graphs combined with multimodal data provides systematic technical support and multi-dimensional research perspectives for oracle bone inscription research. The research results show that the oracle bone inscription knowledge graph plays a significant role in improving the efficiency of oracle bone interpretation, exploring the social relationship network of the Shang Dynasty and promoting the digital protection of traditional culture.