Text Mining Targeted Association API

Description: API serving explicitly targeted Biolink Associations extracted from sentences in the scientific literature. Here, targeted refers to the fact that this service is based on text-mining models targeted to extract specific associations between concepts, as opposed to concepts cooccurring with each other.

Cross-references: https://github.com/NCATSTranslator/Translator-All/wiki/Text%E2%80%90mined-Assertion-KP

Edge Categories Distribution

Epistemic Robustness

Epistemic Robustness measures the provenance quality of edges from this knowledge source, using Knowledge Level and Agent Type from the Biolink Model to assess evidence strength and reliability. Higher scores indicate stronger evidence and greater manual curation.

-0.15
Average Epistemic Score

0.00
Knowledge Level Score
Most common: Not Provided

-0.30
Agent Type Score
Most common: Text-Mining Agent

1,113,708
Edges Included

0
Edges with Missing Provenance
(Both Knowledge Level and Agent Type not provided)

Edge Type Validation

This section shows the validation of edge types in this knowledge source against the recognized biolink schema patterns.

719,867
Recognized Edges
( 64.6 %)

393,763
Unrecognized Edges
( 35.4 %)

ABox / TBox Classification

The TBox-to-ABox balance reflects how much a knowledge graph emphasizes abstract schema versus concrete instances—too much of either can hinder effective learning and reasoning.

1,113,630
ABox Edges
( 100.0 %)

0
TBox Edges
( 0.0 %)

0
Undefined
( 0.0 %)