For the military scenario named entity, the article proposes a supervised named entity recognition method based on deep neural network, which aims to identify and extract the troops, geographical location, weapons and equipment, organization, facilities, battlefield environment, time, etc. in the military scenario. The method avoids the complexity of artificially constructed features and the inaccuracy of military text segmentation. Bi-directional Long Short-Term Memory neural network based on character vector and the conditional random field model are used to automatically extract text features, and then identify the military scenario named entities. Experiments show that the method is higher in recognition accuracy than the traditional method and close to the level of named entity recognition in the 2eneral field.
[1]
Masanori Hattori,et al.
Character-Based LSTM-CRF with Radical-Level Features for Chinese Named Entity Recognition
,
2016,
NLPCC/ICCPOL.
[2]
Jinhong K. Guo,et al.
Extracting Meaningful Entities from Human-generated Tactical Reports
,
2015,
Complex Adaptive Systems.
[3]
Kenny Q. Zhu,et al.
Multi-channel BiLSTM-CRF Model for Emerging Named Entity Recognition in Social Media
,
2017,
NUT@EMNLP.
[4]
Guillaume Lample,et al.
Neural Architectures for Named Entity Recognition
,
2016,
NAACL.
[5]
Jason Weston,et al.
Natural Language Processing (Almost) from Scratch
,
2011,
J. Mach. Learn. Res..