Conference proceeding
Expanding Consumer Health Vocabularies with Frequency-Conserving Internal Context Models
2018 IEEE International Conference on Healthcare Informatics (ICHI), pp 241-246
Jun 2018
Featured in Collection : UN Sustainable Development Goals @ Drexel
Abstract
Consumer Health Vocabularies (CHVs) function as lexicons that help healthcare professionals and consumers communicate effectively regarding medical concepts. A CHV is a record of a list of terms that are used by consumers when discussing health-related issues, as well as the associated medical concepts and terminology. In this work, we describe an algorithm to identify candidate terms and associated concepts for inclusion in the CHV from analyzing user-generated text on internet health forums. The proposed algorithm aims to identify terms in user-generated text that are similar to existing terms in the CHV and identify the closest Universal Medical Language System (UMLS) concept for the candidate terms. The model utilizes internal contexts of phrases to generate a likelihood ranking for each phrase observed in the input data. We demonstrate a limited evaluation of model performance and present a list of candidate terms generated by the model.
Metrics
Details
- Title
- Expanding Consumer Health Vocabularies with Frequency-Conserving Internal Context Models
- Creators
- Munif Ishad Mujib - Drexel UniversityChristopher C Yang - Drexel UniversityMengnan Zhao - Drexel UniversityJake Ryland Williams - Drexel UniversityJoshua Randal Williams - Chemistry
- Publication Details
- 2018 IEEE International Conference on Healthcare Informatics (ICHI), pp 241-246
- Conference
- 2018 IEEE International Conference on Healthcare Informatics (ICHI)
- Publisher
- IEEE
- Number of pages
- 1
- Resource Type
- Conference proceeding
- Language
- English
- Academic Unit
- Information Science; Chemistry
- Web of Science ID
- WOS:000853207500027
- Scopus ID
- 2-s2.0-85051123941
- Other Identifier
- 991019174052704721
UN Sustainable Development Goals (SDGs)
This publication has contributed to the advancement of the following goals:
InCites Highlights
Data related to this publication, from InCites Benchmarking & Analytics tool:
- Web of Science research areas
- Computer Science, Artificial Intelligence
- Computer Science, Information Systems
- Health Care Sciences & Services
- Medical Informatics