BiLSTMCRF Empowered Knowledge Graph of Ideological and Political Education Integration Path of Basic Education Teaching Resources

Under the policy guidance of integrated ideological and political education across primary, secondary, and tertiary schools and the background of digital transformation in education, basic education teaching resources face structural problems such as fragmentation, weak knowledge association, insufficient adaptability, and low integration efficiency. Knowledge graphs, as structured knowledge organization and intelligent retrieval tools, provide a technical path for organizing teaching resources with explicit entity–relationship logic. Similar graph-based resource integration is also important in engineering education, including advanced electromagnetic courses where concepts such as electromagnetic waves, antennas, transmission, and propagation require hierarchical semantic linkage. To reduce the limitations of manual annotation, this study applies a BiLSTM-CRF sequence-labeling model to automatically identify entities and extract relationships from ideological and political texts. Textbooks, lesson plans, policy documents, and teaching cases were collected, preprocessed, and annotated to construct a training dataset. Based on the extracted entity–relationship–attribute triples, a knowledge graph was built and used to support resource matching, intelligent retrieval, and personalized recommendation. Empirical analysis using middle-school Ethics and Rule of Law teaching resources verifies that the proposed model improves knowledge extraction efficiency and resource integration quality. The study provides a scalable technical framework for dynamic teaching-resource construction in humanities and engineering education scenarios.