TY - JOUR
T1 - Unsupervised cross-domain named entity recognition using entity-aware adversarial training
AU - PENG, Qi
AU - ZHENG, Changmeng
AU - CAI, Yi
AU - WANG, Tao
AU - XIE, Haoran
AU - LI, Qing
N1 - Publisher Copyright:
© 2020
PY - 2021/6
Y1 - 2021/6
N2 - The success of neural network based methods in named entity recognition (NER) is heavily relied on abundant manual labeled data. However, these NER methods are unavailable when the data is fully-unlabeled in a new domain. To address the problem, we propose an unsupervised cross-domain model which leverages labeled data from source domain to predict entities in unlabeled target domain. To relieve the distribution divergence when transferring knowledge from source to target domain, we apply adversarial training. Furthermore, we design an entity-aware attention module to guide the adversarial training to reduce the discrepancy of entity features between different domains. Experimental results demonstrate that our model outperforms other methods and achieves state-of-the-art performance.
AB - The success of neural network based methods in named entity recognition (NER) is heavily relied on abundant manual labeled data. However, these NER methods are unavailable when the data is fully-unlabeled in a new domain. To address the problem, we propose an unsupervised cross-domain model which leverages labeled data from source domain to predict entities in unlabeled target domain. To relieve the distribution divergence when transferring knowledge from source to target domain, we apply adversarial training. Furthermore, we design an entity-aware attention module to guide the adversarial training to reduce the discrepancy of entity features between different domains. Experimental results demonstrate that our model outperforms other methods and achieves state-of-the-art performance.
KW - Adversarial training
KW - Entity-aware attention
KW - Named entity recognition
KW - Unsupervised cross-domain
UR - http://www.scopus.com/inward/record.url?scp=85101074222&partnerID=8YFLogxK
U2 - 10.1016/j.neunet.2020.12.027
DO - 10.1016/j.neunet.2020.12.027
M3 - Journal Article (refereed)
C2 - 33631608
SN - 0893-6080
VL - 138
SP - 68
EP - 77
JO - Neural Networks
JF - Neural Networks
ER -