SMedBERT: A Knowledge-Enhanced Pre-trained Language Model with Structured Semantics for Medical Text Mining
This work addresses the challenge of understanding complex medical terms and relations in text for applications in medical text mining, representing an incremental improvement by integrating knowledge into existing PLM frameworks.
The authors tackled the problem of enhancing pre-trained language models for medical text mining by incorporating structured semantic knowledge from knowledge graphs, resulting in SMedBERT which significantly outperforms baselines in Chinese medical tasks and improves performance in tasks like question answering and natural language inference.
Recently, the performance of Pre-trained Language Models (PLMs) has been significantly improved by injecting knowledge facts to enhance their abilities of language understanding. For medical domains, the background knowledge sources are especially useful, due to the massive medical terms and their complicated relations are difficult to understand in text. In this work, we introduce SMedBERT, a medical PLM trained on large-scale medical corpora, incorporating deep structured semantic knowledge from neighbors of linked-entity.In SMedBERT, the mention-neighbor hybrid attention is proposed to learn heterogeneous-entity information, which infuses the semantic representations of entity types into the homogeneous neighboring entity structure. Apart from knowledge integration as external features, we propose to employ the neighbors of linked-entities in the knowledge graph as additional global contexts of text mentions, allowing them to communicate via shared neighbors, thus enrich their semantic representations. Experiments demonstrate that SMedBERT significantly outperforms strong baselines in various knowledge-intensive Chinese medical tasks. It also improves the performance of other tasks such as question answering, question matching and natural language inference.