Multi-level semantic fusion network for Chinese medical named entity recognition

Shi, Jincheng; Sun, Mengxuan; Sun, Zhengya; Li, Mingda; Gu, Y. T.; Zhang, Wensheng

doi:10.1016/j.jbi.2022.104144

Cited by 20 publications

(16 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To achieve this research purpose, we searched for the powerful evidence in the relevant policy documents issued by China’s government. Therefore, we adopted the named entity recognition method [ 20 , 21 ] to identify the network nodes of the HnICN, i.e., the government departments and public organizations mentioned in the relevant documents, and utilized the social network analysis [ 22 , 23 ] to explore the relationships between different nodes.…”

Section: The Healthcare–nursing Information Collaboration Network (Hn...mentioning

confidence: 99%

Promoting the Integration of Elderly Healthcare and Elderly Nursing: Evidence from the Chinese Government

Hao

Yin

2022

IJERPH

View full text Add to dashboard Cite

The increase of the aging population in China and the rise of the concept of healthy aging have accelerated the transformation and upgrading of the traditional elderly nursing pattern. Nevertheless, there is a critical limitation existing in the current situation of China’s elderly care, i.e., the medical institutions do not support elderly nursing and the elderly nursing institutions do not facilitate access to medical care. To eliminate the adverse impact of this issue, twelve ministries and commissions of the Chinese government have jointly issued a document, i.e., the Several Opinions on Further Promoting the Development of Combining the Healthcare with the Elderly care (SOFPDCHE), to provide guidance from the government level for further promoting the integration of elderly healthcare and elderly nursing. Under this background, this paper constructs a healthcare–nursing information collaboration network (HnICN) based on the SOFPDCHE, proposing three novel strategies to explore the different roles and collaboration relationships of relevant government departments and public organizations in this integration process, i.e., the node identification strategy (NIS), the local adjacency subgroup strategy (LASS), and the information collaboration effect measurement strategy (ICEMS). Furthermore, this paper retrieves 484 valid policy documents related to “the integration of elderly healthcare and elderly nursing” as data samples on the official websites of 12 sponsored ministries and commissions, and finally confirms 22 government departments and public organizations as the network nodes based on these obtained documents, such as the National Health Commission of the People’s Republic of China (NHC), the Ministry of Industry and Information Technology of the People’s Republic of China (MIIT), and the National Working Commission on Aging (NWCA). In terms of the collaboration effect, the results of all node-pairs in the HnICN are significantly different, where the collaboration effect between the NHC and MIIT is best and that between the NATCM and MIIT is second best, which are 84.572% and 20.275%, respectively. This study provides the quantifiable results of the information collaboration degree between different government agencies and forms the optimization scheme for the current collaboration status based on these results, which play a positive role in integrating elderly healthcare and elderly nursing and eventually achieving healthy aging.

show abstract

Section: The Healthcare–nursing Information Collaboration Network (Hn...mentioning

confidence: 99%

Promoting the Integration of Elderly Healthcare and Elderly Nursing: Evidence from the Chinese Government

Hao

Yin

2022

IJERPH

View full text Add to dashboard Cite

show abstract

“…For example, predefined types in EMRs may include “疾病 (Disease)”, “治疗 (Treatment)”, and “检查 (Lab Test)”, while those in TCM manuals may include “药品 (Drug)”, “药物成分 (Drug Ingredient)”, and “药物性味 (Drug Taste)”. Due to the broad range of applications associated with the extracted medical entities, such as medical knowledge graph construction [24] , medical relation extraction [38] , and medical question answering [15] , Chinese medical NER has attracted considerable attention [40] , [18] , [27] , [34] . Although Chinese NER has shown promising results in the general domain [39] , [28] , [17] , [10] , it encounters significant challenges when applied in the medical domain.…”

Section: Introductionmentioning

confidence: 99%

“…Recently, radicals of each Chinese character are also exploited to enhance the recognition of medical terms from unstructured medical texts, especially helping the identification of entity types [18] , [27] . For instance, in Fig.…”

Section: Introductionmentioning

confidence: 99%

“…Therefore, the effective incorporation of radical information is crucial for the task. A straightforward solution is to integrate the radical features in the embedding layer [18] , [27] through a concatenation operation. However, this only achieves token-level fusion and does not model the dense interactions between radical and lattice features.…”

Section: Introductionmentioning

confidence: 99%

“…CLART takes advantage of FLAT [17] in modeling lattice features and further extends FLAT by incorporating radical features, which play a guiding role in medical texts. Unlike the straightforward integration of both lattice and radical features by concatenation operation [18] , [27] , CLART employs a radical Transformer to effectively model the dense interactions between the lattice and radical features, thus facilitating better fusion of the lexical and radical knowledge. Specifically, given a sentence, a medical lexicon, and a radical dictionary, we construct a flat lattice ( i.e., character-word sequence or token sequence) for the sentence and radical components of each Chinese character through word matching and radical parsing, respectively.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

CLART: A cascaded lattice-and-radical transformer network for Chinese medical named entity recognition

Xiao,

Ji,

et al. 2023

Heliyon

View full text Add to dashboard Cite

Dynamic Multi-View Fusion Mechanism for Chinese Relation Extraction

Yang

et al. 2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Recently, many studies incorporate external knowledge into character-level feature based models to improve the performance of Chinese relation extraction. However, these methods tend to ignore the internal information of the Chinese character and cannot filter out the noisy information of external knowledge. To address these issues, we propose a mixture-of-view-experts framework (MoVE) to dynamically learn multi-view features for Chinese relation extraction. With both the internal and external knowledge of Chinese characters, our framework can better capture the semantic information of Chinese characters. To demonstrate the effectiveness of the proposed framework, we conduct extensive experiments on three real-world datasets in distinct domains. Experimental results show consistent and significant superiority and robustness of our proposed framework. Our code and dataset will be released at: https://gitee.com/tmg-nudt/multi-view-of-expert-for-chinese-relation-extraction

show abstract

Multi-level semantic fusion network for Chinese medical named entity recognition

Cited by 20 publications

References 29 publications

Promoting the Integration of Elderly Healthcare and Elderly Nursing: Evidence from the Chinese Government

Promoting the Integration of Elderly Healthcare and Elderly Nursing: Evidence from the Chinese Government

CLART: A cascaded lattice-and-radical transformer network for Chinese medical named entity recognition

Dynamic Multi-View Fusion Mechanism for Chinese Relation Extraction

Contact Info

Product

Resources

About