Comparing CNN and LSTM character-level embeddings in BiLSTM-CRF models for chemical and disease named entity recognition

2018/08/25 by Zenan Zhai, Zhai, Zenan, Dat Quoc Nguyen +3
Computer Science · #Computation and Language (cs.CL) #FOS: Computer and information sciences #cs.CL

paper · pdf · doi:10.48550/arxiv.1808.08450

In Proceedings of the 9th International Workshop on Health Text Mining and Information Analysis (LOUHI 2018), to appear

arxiv created 2018/08/25 · arxiv updated 2018/08/28

Abstract

We compare the use of LSTM-based and CNN-based character-level word embeddings in BiLSTM-CRF models to approach chemical and disease named entity recognition (NER) tasks. Empirical results over the BioCreative V CDR corpus show that the use of either type of character-level word embeddings in conjunction with the BiLSTM-CRF models leads to comparable state-of-the-art performance. However, the models using CNN-based character-level word embeddings have a computational performance advantage, increasing training time over word-based models by 25% while the LSTM-based character-level word embeddings more than double the required training time.

Related

• Convolutional neural networks for chemical-disease relation extraction are improved with character-based word embeddings
• Unified Neural Architecture for Drug, Disease and Clinical Entity Recognition
• Supervised Fine-Tuning or In-Context Learning? Evaluating LLMs for Clinical NER
• Integrating Dictionary Feature into A Deep Learning Model for Disease Named Entity Recognition
• Multi-level biomedical NER through multi-granularity embeddings and enhanced labeling