Incorporating Domain Knowledge into Medical NLI using Knowledge Graphs

Abstract

Recently, biomedical version of embeddings obtained from language models such as BioELMo have shown state-of-the-art results for the textual inference task in the medical domain. In this paper, we explore how to incorporate structured domain knowledge, available in the form of a knowledge graph (UMLS), for the Medical NLI task. Specifically, we experiment with fusing embeddings obtained from knowledge graph with the state-of-the-art approaches for NLI task (ESIM model). We also experiment with fusing the domain-specific sentiment information for the task. Experiments conducted on MedNLI dataset clearly show that this strategy improves the baseline BioELMo architecture for the Medical NLI task.

EMNLP 2019 accepted short paper

References in corpus (3)

• BioBERT: a pre-trained biomedical language representation model for biomedical text mining
• Neural Natural Language Inference Models Enhanced with External Knowledge
• Probing Biomedical Embeddings from Language Models

Cited by in corpus (7)

• Knowledge-based Biomedical Data Science 2019
• Pre-trained Language Models in Biomedical Domain: A Systematic Survey
• MedSyn: LLM-based Synthetic Medical Text Generation Framework
• Infusing Disease Knowledge into BERT for Health Question Answering, Medical Inference and Disease Name Recognition
• Joint Biomedical Entity and Relation Extraction with Knowledge-Enhanced Collective Inference
• Probing Pre-Trained Language Models for Disease Knowledge
• Box Embeddings: An open-source library for representation learning using geometric structures