Comparing CNN and LSTM character-level embeddings in BiLSTM-CRF models for chemical and disease named entity recognition
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.
Keywords
Cite
@article{arxiv.1808.08450,
title = {Comparing CNN and LSTM character-level embeddings in BiLSTM-CRF models for chemical and disease named entity recognition},
author = {Zenan Zhai and Dat Quoc Nguyen and Karin Verspoor},
journal= {arXiv preprint arXiv:1808.08450},
year = {2018}
}
Comments
In Proceedings of the 9th International Workshop on Health Text Mining and Information Analysis (LOUHI 2018), to appear