Exploring Partial Knowledge Base Inference in Biomedical Entity Linking
This work addresses a practical but understudied scenario in biomedical entity linking that is incremental, focusing on improving robustness for stakeholders with partial knowledge bases.
The paper tackles the problem of biomedical entity linking when only a subset of the knowledge base is available during inference, revealing catastrophic performance degradation due to unlinkable mentions, and proposes two simple methods to address this issue with little computational overhead.
Biomedical entity linking (EL) consists of named entity recognition (NER) and named entity disambiguation (NED). EL models are trained on corpora labeled by a predefined KB. However, it is a common scenario that only entities within a subset of the KB are precious to stakeholders. We name this scenario partial knowledge base inference: training an EL model with one KB and inferring on the part of it without further training. In this work, we give a detailed definition and evaluation procedures for this practically valuable but significantly understudied scenario and evaluate methods from three representative EL paradigms. We construct partial KB inference benchmarks and witness a catastrophic degradation in EL performance due to dramatically precision drop. Our findings reveal these EL paradigms can not correctly handle unlinkable mentions (NIL), so they are not robust to partial KB inference. We also propose two simple-and-effective redemption methods to combat the NIL issue with little computational overhead. Codes are released at https://github.com/Yuanhy1997/PartialKB-EL.