Biomedical named entity recognition using deep neural networks with contextual information
Biomedical named entity recognition using deep neural networks with contextual information
复制标题
DOI：
10.1186/s12859-019-3321-4
复制
发表时间：
2019-12-30
影响因子：
3
通讯作者：
Lee, Hyunju
中科院分区：
文献类型：
--
作者：
Cho, Hyejin;Lee, Hyunju
背景：在生物医学文本挖掘中，命名实体识别（NER）是用于从生物医学文章中提取信息的重要任务。之前提出的 NER 方法是基于字典或规则的方法和机器学习方法。然而，这些传统方法严重依赖于大规模词典、针对特定目标的规则或结构良好的语料库。这些 NER 方法已被独立于手工特征的基于深度学习的方法所取代。然而，尽管此类 NER 方法采用额外的条件随机场 (CRF) 来捕获相邻标签之间的重要相关性，但它们通常不会将文本中的所有上下文信息合并到深度学习层中。该系统被称为具有 CRF 的上下文长短期记忆网络（CLSTM）。我们在三个语料库上评估 CLSTM 模型：国家生物技术信息中心 (NCBI) 的疾病语料库、BioCreative II 基因提及语料库 (GM) 和 BioCreative V 化学疾病关系语料库 (CDR)。我们的框架与几种深度学习方法进行了比较，例如 BiLSTM、BiLSTM with CRF、GRAM-CNN 和 BERT。在 NCBI 语料库上，我们的模型记录了疾病 NER 的 F 分数为 85.68%，比之前的方法提高了 1.50%。此外，尽管 BERT 通过合并超过 25 亿个单词来使用迁移学习，但我们的系统表现出与 BERT 相似的性能，在 GM 语料库上的基因 NER 的 F 分数为 81.44%，在 CDR 语料库上的化学品和疾病的 NER 的 F 分数为 86.44%。我们得出的结论是，我们的方法显着提高了生物医学 NER 任务的性能。结论：所提出的方法在识别文本中的生物实体方面非常稳健。
Background: In biomedical text mining, named entity recognition (NER) is an important task used to extract information from biomedical articles. Previously proposed methods for NER are dictionary- or rule-based methods and machine learning approaches. However, these traditional approaches are heavily reliant on large-scale dictionaries, target-specific rules, or well-constructed corpora. These methods to NER have been superseded by the deep learning-based approach that is independent of hand-crafted features. However, although such methods of NER employ additional conditional random fields (CRF) to capture important correlations between neighboring labels, they often do not incorporate all the contextual information from text into the deep learning layers.Results: We propose herein an NER system for biomedical entities by incorporating n-grams with bi-directional long short-term memory (BiLSTM) and CRF; this system is referred to as a contextual long short-term memory networks with CRF (CLSTM). We assess the CLSTM model on three corpora: the disease corpus of the National Center for Biotechnology Information (NCBI), the BioCreative II Gene Mention corpus (GM), and the BioCreative V Chemical Disease Relation corpus (CDR). Our framework was compared with several deep learning approaches, such as BiLSTM, BiLSTM with CRF, GRAM-CNN, and BERT. On the NCBI corpus, our model recorded an F-score of 85.68% for the NER of diseases, showing an improvement of 1.50% over previous methods. Moreover, although BERT used transfer learning by incorporating more than 2.5 billion words, our system showed similar performance with BERT with an F-scores of 81.44% for gene NER on the GM corpus and a outperformed F-score of 86.44% for the NER of chemicals and diseases on the CDR corpus. We conclude that our method significantly improves performance on biomedical NER tasks.Conclusion: The proposed approach is robust in recognizing biological entities in text.