Skip to main content

Table 2 Evaluation against RegulonDB

From: Improving the extraction of complex regulatory events from scientific text by using ontology-based inference

Corpus Recall Precision F-score
ecoli-tf.abstract 22.4% (0.3%) 77.2% (50.0%) 0.35 (0.01)
ecoli-tf.fulltext 24.0% (1.5%) 67.1% (76.1%) 0.35 (0.03)
regulon.abstract 17.1% (0.1%) 85.0% (80.0%) 0.28 (0.00)
regulon.fulltext 14.1% (1.2%) 74.0% (91.7%) 0.24 (0.02)
Total 33.1% (2.1%) 66.6% (79.6%) 0.44 (0.04)