Sub-Term Mapping Tools

All Terms with no SMT synonyms lsit

The setup of thsi test are described as follows:

  • Input Terms: all terms from UMLS-Core:
    • Original ID|TERM|CUI: 15,447
    • Total term with multi-CUI: 35
    • Total term with multi-ID: 1,492
    • It has been modified to 13,077 input TERM|CUI.

  • SMT model:
    • Lexicon, 2013
    • Meta-thesaurus, 2013AA
    • Synonyms list: none

Test IDSynonymslvg flowNo Sub.1 Sub.2 Sub.PrecisionRecallF1-measureMetaMap
Distance
Score
baselineNone 78.23% (10,230)0.0% (0)0.00% (0)64.49%72.02%0.6805 
sspelling variants-f:s78.23% (10,230)0.30% (39)0.01% (1)64.44%72.24%0.68120
iinflectional variants-f:i78.23% (10,230)0.02% (3)0.00% (0)64.48%72.03%0.68051
ysynonyms-f:y78.23% (10,230)0.18% (23)0.00% (0)64.43%72.10%0.68052
AAcronym/Abbreviation variants-f:A78.23% (10,230)3.04% (398)0.13% (17)64.18%75.41%0.69342
aexpansion variants-f:a78.23% (10,230)3.12% (408)0.17% (22)64.17%75.48%0.69362
dderivational variants-f:d78.23% (10,230)0.83% (109)0.04% (5)64.40%72.72%0.68313
Gefruitful variants-f:Ge78.23% (10,230)4.60% (602)0.39% (51)63.20%76.75%0.6932varies
SMTSMT synonyms 78.23% (10,230)4.74% (620)0.31% (40)63.48%77.52%0.6980N/A

  • Precision = relevant, retrieved / total retrieved
  • Recall = relevant, retrieved / total relevant
  • F1 = (2 x Precision x Recall) / (Precision + Recall)
    precision and recall are equally important, so use F1 (β = 1)