Chemlistem: chemical named entity recognition using recurrent neural networks
Chemlistem: chemical named entity recognition using recurrent neural networks
复制标题
DOI：
10.1186/s13321-018-0313-8
复制
发表时间：
2018-12-06
影响因子：
8.6
通讯作者：
Boyle, John
中科院分区：
文献类型：
--
作者：
Corbett, Peter;Boyle, John
化学命名实体识别（NER）传统上由基于条件随机场（CRF）的方法主导，但鉴于被称为深度学习的人工神经网络技术的成功，我们决定研究它们作为CRF的替代方案。本文介绍了几种化学命名实体识别系统。第一个系统将传统的基于crf的习语转换为深度学习框架，使用丰富的每标记特征和神经词嵌入，并使用双向长短期记忆（LSTM）网络（一种循环神经网络）产生一系列标签。第二个系统避免了丰富的特征集甚至标记化，转而使用神经特征嵌入和多个LSTM层进行字符标记。第三个系统是前两个系统结果的综合。我们最初的BioCreative V.5参赛作品被评为F分最高的第一组，随后使用迁移学习在测试数据上获得了90.33%的最终F分（准确率91.47%，召回率89.21%）。
Chemical named entity recognition (NER) has traditionally been dominated by conditional random fields (CRF)-based approaches but given the success of the artificial neural network techniques known as deep learning we decided to examine them as an alternative to CRFs. We present here several chemical named entity recognition systems. The first system translates the traditional CRF-based idioms into a deep learning framework, using rich per-token features and neural word embeddings, and producing a sequence of tags using bidirectional long short term memory (LSTM) networksa type of recurrent neural net. The second system eschews the rich feature setand even tokenisationin favour of character labelling using neural character embeddings and multiple LSTM layers. The third system is an ensemble that combines the results of the first two systems. Our original BioCreative V.5 competition entry was placed in the top group with the highest F scores, and subsequent using transfer learning have achieved a final F score of 90.33% on the test data (precision 91.47%, recall 89.21%).