UMLS Knowledge Graph

Directed knowledge graph from the Unified Medical Language System semantic network. Nodes are biomedical concepts and edges are semantic relations (e.g. 'treats', 'diagnoses', 'prevents'). Multiple relation types can connect the same concept pair, making it a directed multigraph. 46 relation types serve as edge attributes.

npx keepset fetch kg-umls

Nodes: 135
Edges: 6,529
Size: 34.8 KiB

Fetch and preview this dataset on this page, or download the result as JSON. The command line fetches it without a browser.

Overview

Id	`kg-umls`
Category	`graph/v1`
Kind	graph
Domain	Biomedical knowledge graph
Tags	knowledge-graph
Licence	MIT
Catalogue	`keepset` 1.33.0

Sources and provenance

#	URL	Status	Retrieved	Read by recipe	Pinned snapshot	SHA-256
0	https://github.com/TimDettmers/ConvE/raw/master/umls.tar.gz	live	2026-04-19	yes

Citations

S. Kok, P. Domingos (2007). Statistical Predicate Invention. Proceedings of the 24th International Conference on Machine Learning, 433-440.

Recipe

Source	#0
Archive	tar.gz
Part train	`train.txt`
Part valid	`valid.txt`
Part test	`test.txt`
Parser format	`"kg-triples"`
Parser encoding	`"utf-8"`
Parser delimiter	`"tab"`
Parser order	`"head-relation-tail"`
Parser header	`"none"`
Parser quoted	`false`

Crosswalk

keepset/relations-split to graph/v1, as a crossmap crosswalk.

Rule	From	To	Direction
move	`meta`	`meta`	both
move	`nodes[].id`	`nodes[].id`	both
move	`edges[].source`	`edges[].source`	both
move	`edges[].target`	`edges[].target`	both
move	`edges[].type`	`edges[].type`	both
move	`edges[].split`	`edges[].split`	both

Expected counts

Edges	6,529
Nodes	135

Properties

Directed	yes
Download bytes	34.8 KiB
Heterogeneous	yes
Multigraph	yes
Self loops	no
Weighted	no