arrow
Return

Biomedical entity linking based on less labeled data

delete2022-04-05
delete2
PRE
AI
H
Hu, Yu *
D
Derong Shen
T
Tiezheng Nie
Y
Yue Kou
G
Ge Yu
DOI:10.1007/s11704-022-1192-8delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
Conclusion In this work, we perform a joint learning model to construct a fine-grained biomedical entity and relation linking. We adopt a probabilistic a probabilistic logic network to align the entities in biomedical text to the concepts in the knowledge base. And we utilize the dispersion degree of standard deviation to evaluate feature distribution and select those tuples with distinguishing features for interaction relationship classifying. Then we rank the candidate entity-relation triples to obtain the linking result.

Journal

Frontiers of Computer Science cover
Frontiers of Computer Science
IF:
4.6
Papers:
1.6K
Citations:
2.8K

Organization

N
northeastern university - china
Scholars:
3.1W
Papers: 2.7W
Citations: 37