A multitask bi-directional RNN model for named entity recognition on Chinese electronic medical records

BMC Bioinformatics

Table 1 The proposed network architecture

Name	Description
Input	Sentences in EMR
Word Embedding	Mikolov model
Character Embedding Layer	150 LSTM cells for each hidden layer,
	one forward hidden layer and one backward hidden layer,
	Dropout = 0.5
Parts-of-speech tag (POS) layer	150 LSTM cells for each hidden layer,
	one forward hidden layer and one backward hidden layer,
	Dropout = 0.5
Named Entity recognition (NER) Layer	150 LSTM cells for each hidden layer,
	one forward hidden layer and one backward hidden layer,
	Dropout = 0.5
Output	Softmax

ISSN: 1471-2105