Skip to main content

Table 6 Average speed in CUI concept pairs per second (pairs/s) for the evaluation of random CUI pairs with three representative ontology-based similarity measures based on the SNOMED-CT US 2019AB ontology (357,406 nodes) implemented by the three UMLS-based semantic measures libraries reported in the literature

From: HESML: a real-time semantic measures library for the biomedical domain with a reproducible survey

Similarity measure

UMLS::Similarity

SML

HESML

 

Avg. speed (pairs/s)

Avg. speed (pairs/s)

Avg. speed (pairs/s)

Rada [71]

0.122 (15)

xxx

0.041 (15)

AncSPL-Rada

(this work)

–

–

30110 \((10^7)\)

Lin-Seco [87, 110]

0.744 (500)

202160 \((10^7)\)

491942 \((10^7)\)

Wu-Palmer\(_{fast}\) [72]

0.035 (15)

–

435252 \((10^7)\)

  1. Best performing values are shown in bold. Non-implemented methods (–) or more than 1 h/pair (xxx). UMLS::Similarity uses caching for the shortest path computations. The number of random CUI pairs evaluated to measure each value is shown between parentheses
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us