Incorporating Domain Knowledge into Medical NLI using Knowledge Graphs
This work addresses the challenge of enhancing inference accuracy in the medical domain for applications like clinical decision support, though it is incremental as it builds on established methods.
The paper tackled the problem of improving medical natural language inference by incorporating structured domain knowledge from UMLS knowledge graphs and domain-specific sentiment information into existing models, resulting in improved performance over the BioELMo baseline on the MedNLI dataset.
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.