From electronic health records to terminology base: A novel knowledge base enrichment approach

Jiaying Zhang, Zhixing Zhang, Huanhuan Zhang, Zhiyuan Ma, Qi Ye, Ping He, Yangming Zhou

Abstract: Highlights•We propose S2ERP to align a set of external terms to an existing TB.•It integrates BERT-based semantic embedding and GCN-based structure embedding.•BERT hot-starts KB embedding as well as enhances semantic information.•GCN merges neighboring synonyms and hyponyms information into formal entities in TB.•Experimental results show that S2ERP achieves better performance than baselines.