Laura Zanella (Orpailleur) will defend her thesis, entitled “Biomedical Event Extraction Based on Transformers and Knowledge Graphs“, on Friday, December 15th at 2 pm in room B013.
Abstract :
Biomedical event extraction can be divided into three main subtasks; (1) event trigger detection, (2) argument identification, and (3) event construction. In this work, for the first subtask, we analyze a set of transformer language models that are commonly used in the biomedical domain to evaluate and compare their capacity for event trigger detection. We fine-tune the models using seven manually annotated corpora to assess their performance in different biomedical subdomains. SciBERT emerged as the highest-performing model, presenting a slight improvement compared to baseline models. For the second subtask, we construct a knowledge graph (KG) from the biomedical corpora and integrate its KG embeddings to SciBERT to enrich its semantic information. We demonstrate that adding the KG embeddings to the model improves the argument identification performance by around 20 %, and by around 15 % compared to two baseline models. For the third subtask, we use the generative model, ChatGPT, based on prompts to construct the final set of extracted events. Our results suggest that fine-tuning a transformer model that is pre-trained from scratch with biomedical and general data allows to detect event triggers and identify arguments covering different biomedical subdomains, and therefore improving its generalization. Furthermore, the integration of KG embeddings into the model can significantly improve the performance of biomedical event argument identification, outperforming the results of baseline models.
Jury
Reviewers:
-
Claire Gardent (D.R. CNRS, Université de Lorraine)
-
Mathieu Roche (D.R. CIRAD)
-
Thierry Charnois (Pr. Université de Paris 13)
Thesis Director:
- Yannick Toussaint (Pr. Université de Lorraine)
