Dataset Open Access
CORD-19 Diachronic Concept Embeddings
Newman-Griffis, Denis;
Sivaraman, Venkatesh;
Fosler-Lussier, Eric;
Perer, Adam;
Hochheiser, Harry
Medical concept embeddings trained on monthly releases of the CORD-19 dataset, for analysis of diachronic change.
This resource includes:
- Concept embeddings, trained using the JET toolkit on the set of articles added to CORD-19 in each month of March 2020 - October 2020. Concepts are SNOMED CT codes.
- Nearest neighborhood-based analysis results of the concept embeddings, stored as a SQLite database.
Database content excludes terms and definitions for SNOMED CT codes, which are licensed content owned by SNOMED International. Information on accessing SNOMED licensed content may be found here.
Files
(1.3 GB)
| Name | Size | |
|---|---|---|
|
CORD-19_analysis__2020-03__2020-10.db
md5:5bc7c50f6bfd89c1c33d34d394556c92 |
340.3 MB | Download |
|
CORD-19_monthly_embeddings.zip
md5:be0737c6c04a9e51fc04e94f384dc1f9 |
998.5 MB | Download |
5,834
509
views
downloads
| All versions | This version | |
|---|---|---|
| Views | 5,834 | 2,626 |
| Downloads | 509 | 132 |
| Data volume | 85.5 GB | 63.3 GB |
| Unique views | 5,325 | 2,452 |
| Unique downloads | 342 | 102 |
Indexed in
- Publication date:
- February 14, 2021
- DOI:
-
- Related identifiers:
- References
- 10.18653/v1/W18-3026 (Conference paper)
- Communities:
- License (for files):
- Creative Commons Attribution 4.0 International