Data mining (DM) technology is increasingly used in higher education, especially imaging education. The IEKMM model connects knowledge, problems, and abilities, addressing asymmetrical relationships and supporting network reasoning tasks. The SSME model preserves IEKMM’s semantic information, enhancing instruction quality and efficiency, and advancing personalized learning initiatives. Findings reveal that the distributed representation of entities and relationships, trained using the SSME (Semantic Symbol Mapping Embedding) model, effectively preserves the original semantic information of the IEKMM. This provides a foundation for implementing knowledge maps in educational settings and is crucial for advancing personalized learning initiatives.