- Title Information
- Title
- Unsupervised induction of domain-specific semantic frames from operative notes for surgical clinical research
- Name:
Personal
- Name Part
- Wang, Yan
- Role
- Role Term:
Text (marcrelator)
- author
- Name:
Personal
- Name Part
- Pakhomov, Serguei
- Role
- Role Term:
Text (marcrelator)
- author
- Name:
Personal
- Name Part
- Melton, Genevieve B.
- Role
- Role Term:
Text (marcrelator)
- author
- Type of Resource
- text
- Genre (Local)
- poster abstracts
- Origin Information
- Date Created
(keyDate="yes", encoding="w3cdtf")
- 2015
- Language
- Language Term:
Code (ISO639-2B)
- eng
- Note
(displayLabel="Source")
- Electronic Health Record
- Note
(displayLabel="Method")
- Natural Language Processing
- Abstract
- Surgical clinical research often utilizes information within operative notes including data about the supplies, tools, and techniques used for a given procedure. We have previously demonstrated that compared to general English semantic frames, surgical operative notes contain semantic frames with additional or different arguments and that additional verb senses. In this study, we investigated the use of an unsupervised approach for inducing semantic frames for surgical action verbs. Our results demonstrate that semantic frames induced from domain corpus using unsupervised approach cover most of the domain semantic frames created by human experts
- Note:
funding
- Support for this research came from American Surgical Association Foundation and National Institutes of Health (#R01LM011364(GM), #R01GM102282(SP), #8UL1TR000114).
- Note
- Poster abstract for the 2015 AMIA Annual Symposium
- Related Item:
Host
(displayLabel="Published in")
- Title Information
- Title
- AMIA Joint Summits on Translational Science Proceedings
- Origin Information
- Date Created
- 2015
- Date Created
(keyDate="yes", encoding="w3cdtf")
- 2015
- Physical Description
- Extent
- p. 415-416
- Access Condition:
use and reproduction
(href="https://creativecommons.org/licenses/by-nc/4.0/")
- This work is licensed under a
Creative Commons Attribution-NonCommercial 4.0 International License
- Access Condition:
logo
(href="https://i.creativecommons.org/l/by-nc/4.0/88x31.png")