CCRFs-NER: Named Entity Recognition Method Based on Cascaded Conditional Random Fields Oriented Chinese EMR

Tenth International Conference on Applications and Techniques in Cyber Intelligence (ICATCI 2022)(2023)

Cited 1|Views9
No score
Abstract
Named Entity Recognition(NER) for Chinese Electronic Medical Record (EMR) is to identify medical entities and entity boundaries. Nested entities exist in Chinese electronic medical records, which makes NER low accuracy. To improve the recognition accuracy, this paper proposes a NER method, named CCRFs-NER. Firstly, the feature set is constructed by word features, part-of-speech features, and entity-identifier features. Then, Bi-LSTM and attention mechanism are used to process the feature set to extract the global and local features. Finally, the features are input into the CCRFs for probability prediction, and the recognition results are obtained. The experimental results show that the CCRFs-NER can improve the accuracy of named entity recognition for Chinese electronic medical records.
More
Translated text
Key words
Chinese EMR, Named Entity Recognition, Nested entities, Attention mechanism, CCRFs
AI Read Science
Must-Reading Tree
Example
Generate MRT to find the research sequence of this paper
Chat Paper
Summary is being generated by the instructions you defined