Biomedical Event Extraction with Hierarchical Knowledge Graphs
This work addresses the challenge of extracting nested structured events with non-indicative trigger words in biomedical texts, which is incremental as it builds on existing methods by integrating hierarchical knowledge graphs.
The paper tackled the problem of biomedical event extraction by incorporating domain knowledge from UMLS into a pre-trained language model using Graph Edge-conditioned Attention Networks and hierarchical graph representation, achieving improvements of 1.41% F1 on all events and 3.19% F1 on complex events on the BioNLP 2011 GENIA Event Extraction task.
Biomedical event extraction is critical in understanding biomolecular interactions described in scientific corpus. One of the main challenges is to identify nested structured events that are associated with non-indicative trigger words. We propose to incorporate domain knowledge from Unified Medical Language System (UMLS) to a pre-trained language model via Graph Edge-conditioned Attention Networks (GEANet) and hierarchical graph representation. To better recognize the trigger words, each sentence is first grounded to a sentence graph based on a jointly modeled hierarchical knowledge graph from UMLS. The grounded graphs are then propagated by GEANet, a novel graph neural networks for enhanced capabilities in inferring complex events. On BioNLP 2011 GENIA Event Extraction task, our approach achieved 1.41% F1 and 3.19% F1 improvements on all events and complex events, respectively. Ablation studies confirm the importance of GEANet and hierarchical KG.