Proceedings of the 2015 International Conference on Electromechanical Control Technology and Transportation

A Method of Automatic Annotation for Medical Record Text Based on Latent Dirichlet Allocation

Authors
Xinyu Jin, Qiliang Jin, Yuze Li
Corresponding Author
Xinyu Jin
Available Online November 2015.
DOI
10.2991/icectt-15.2015.58How to use a DOI?
Keywords
medical record text; semantic analysis; Latent Dirichlet Allocation; BM25
Abstract

With the rapid development of medical information, medical data, especially medical record text, are difficult to intelligent analyses, because these data have loose grammar structure. Latent semantic analysis technology in the field of text mining in recent years made extensive research and application, and Latent Dirichlet Allocation(LDA), put forward by Blei, is a method to solve those difficulties. This paper proposed an improved LDA based on BM25 mixture weights method to analyze Chinese medical record text and had a good performance.

Copyright
© 2015, the Authors. Published by Atlantis Press.
Open Access
This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

Volume Title
Proceedings of the 2015 International Conference on Electromechanical Control Technology and Transportation
Series
Advances in Engineering Research
Publication Date
November 2015
ISBN
978-94-6252-124-7
ISSN
2352-5401
DOI
10.2991/icectt-15.2015.58How to use a DOI?
Copyright
© 2015, the Authors. Published by Atlantis Press.
Open Access
This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Cite this article

TY  - CONF
AU  - Xinyu Jin
AU  - Qiliang Jin
AU  - Yuze Li
PY  - 2015/11
DA  - 2015/11
TI  - A Method of Automatic Annotation for Medical Record Text Based on Latent Dirichlet Allocation
BT  - Proceedings of the 2015 International Conference on Electromechanical Control Technology and Transportation
PB  - Atlantis Press
SP  - 305
EP  - 308
SN  - 2352-5401
UR  - https://doi.org/10.2991/icectt-15.2015.58
DO  - 10.2991/icectt-15.2015.58
ID  - Jin2015/11
ER  -