Ruxuan Ou
ABSTRACT
For the named entity recongnition task, with an uneven distribution of entities, the model usually fails to correctly identify minor entities because of the influence of other major entities Therefore, a name entity recognition model based on ensemble column-wise convolution is proposed. Column-wise convolution performs the corresponding convolution operation on each dimension of the word embedding separately, and then concatenates the results into a final feature. Combined with the integrated network architecture, it enables access to richer semantic information to further enhance the generalisation of the model. Experiments were conducted on CMeEE datasets and the model reached 69.47% levels in F1-score, confirming that they were both better than the other comparison models., and there is a significant improvement in the accuracy of recognition of minor entities.
- Li H, Hagiwara M, Li Q, Comparison of the impact of word segmentation on name tagging for Chinese and Japanese[C]//Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC'14). 2014: 2532-2536.Google Scholar
- Collobert R, Weston J, Bottou L, Natural language processing (almost) from scratch[J]. Journal of machine learning research, 2011, 12(ARTICLE): 2493− 2537.Google Scholar
- Huang Z, Xu W, Yu K. Bidirectional LSTM-CRF models for sequence tagging [J]. arXiv preprint arXiv:1508.01991, 2015.Google Scholar
- Vaswani A, Shazeer N, Parmar N, Attention is all you need [J]. Advances in neural information processing systems, 2017, 30.Google Scholar
- WANG Z N, JIANG M, GAO J L, Chinese named entity recognition method based on BERT [J]. Computer Science, 2019, 46 (Z2): 138-142.Google Scholar
- LI N, GUAN H M, YANG P, BERT-IDCNN-CRF for named entity in Chinese [J]. Journal of Shandong University (Natural Science), 2020, 55(01): 102-109.Google Scholar
- Devlin J, Chang M W, Lee K, Bert: Pre-training of deep bidirectional transformers for language understanding [C]//Proceedings of NAACL-HLT. 2019: 4171-4186.Google Scholar
- Sun Y, Wang S, Li Y, Ernie: Enhanced representation through knowledge integration [J]. arXiv preprint arXiv: 1904.09223, 2019.Google Scholar
- TANG K, WANG H T, JIANG Y. Research on opinion target extraction model of e-commerce review based on deep learning [J]. Electronic Measurement Technology, 2019, 42(13): 86-91.Google Scholar
- YUAN Y, WANG Y L, LIU K. Named entity recognition model based on dilated convolutional block architecture [J/OL]. Journal of Shandong University (Engineer Science): 1-10, 2022, 06, 11. http://kns.cnki.net/kcms/detail/37.1391.T.20220323.1735.008.html.Google Scholar
- XING Z W, DAI Z, LUO Q. Combined ECNN and attention mechanism for named entity recognition in civil aviation business [J]. Computer Engineering and Design, 2022, 43(02): 443-449.Google Scholar
- Strubell E, Verga P, Belanger D, Fast and accurate entity recognition with iterated dilated convolutions [J]. arXiv preprint arXiv:1702.02098, 2017.Google Scholar
- Xiong R, Yang Y, He D, On layer normalization in the transformer architecture [C]// International Conference on Machine Learning. PMLR, 2020: 10524-10533.Google Scholar
- Hongying Z, Wenxin L, Kunli Z, Building a pediatric medical corpus: Word segmentation and named entity annotation [C]// Workshop on Chinese Lexical Semantics. Springer, Cham, 2020: 652-664.Google Scholar
Index Terms
- Chinese medical named entity recognition based on ensemble column-wise convolution
Recommendations
Learning multilingual named entity recognition from Wikipedia
We automatically create enormous, free and multilingual silver-standard training annotations for named entity recognition (ner) by exploiting the text and structure of Wikipedia. Most ner systems rely on statistical models of annotated data to identify ...
Two-stage approach to named entity recognition using Wikipedia and DBpedia
IMCOM '17: Proceedings of the 11th International Conference on Ubiquitous Information Management and CommunicationIn natural language understanding, extraction of named entity (NE) mentions in given text and classification of the mentions into pre-defined NE types are important processes. Most NE recognition (NER) relies on resources such as a training corpus or NE ...
Fine-grained Dutch named entity recognition
This paper describes the creation of a fine-grained named entity annotation scheme and corpus for Dutch, and experiments on automatic main type and subtype named entity recognition. We give an overview of existing named entity annotation schemes, and ...
Comments