What question did this study set out to answer?

The aim is to evaluate the effectiveness of systems based on BERT for named entity recognition in low-resource biomedical contexts.

April 1, 2026Open Access

AMI Team at the NTCIR-16 Real-MedNLP Task

Key Points

The aim is to evaluate the effectiveness of systems based on BERT for named entity recognition in low-resource biomedical contexts.
Participated in subtasks 1 and 2 of NTCIR-16 Real-MedNLP Task.
Developed two systems using UTH-BERT pretrained on biomedical documents.
Used ensemble method with hidden vectors from multiple UTH-BERT layers and fine-tuning with CRF layer.
Constructed multistage method for named entity identification with candidate extraction, identification, and tag correction stages.
Systems showed effectiveness in low-resource settings using BERT-based approaches.
Preliminary experiments indicated strong performance of UTH-BERT for named entity recognition.

Abstract

The AMI team participated in subtasks 1 and 2 of the NTCIR-16 Real-MedNLP Task. In this paper, we report our systems employed for subtasks 1 and 2. In subtask 1, the organizer provides a small amount of training data. In recent years, the approach based on BERT has achieved excellent results for such a low-resource situation. We construct two systems based on the BERT model pretrained on biomedical documents (UTH-BERT). We construct the ensemble method with hidden vectors from multiple layers of UTH-BERT and the fine-tuning method with the CRF layer. In subtask 2, participants construct their methods based on the annotation guideline. We construct a multistage method to identify named entities. The system consists of three stages: a candidate extraction stage, an identification stage, and a tag correction stage. We discuss the effectiveness of our systems on the basis of our preliminary experiments and the results in the formal run.

Read Full Paperexternally

اسأل الذكاء الاصطناعي

Bookmark

View Full Paper