Generating Synonyms from STMT Synonyms (UMLS-Core) - TBD

I. Introduction

TBD
The synonym file (normTermSynonyms.data) does not include EUI information. Also, they are normalized. It need more work to change the format and add information.

II. Process

  • Directory: ${LEXICON_SYNONYMS}
  • program: ./Synonym/GenSynonymFromStmt.java
  • Inputs:
    • ./inData/synonyms.data (from Stmt)
    • ./outData/Results/synonymFromMeta.data
    • ./outData/Results/synonymFromNom.data
    • ./outData/Results/synonymFromLvg.data
  • Algorithm:
    • Go through 8 source files
      File NameNotes
      acronym_edited.txtNot used
      greco-latin.txtNot used
      To_add_UMLS_syn_consolidated.txtNot used
      british.txtNot used
      lvg.txtNot used
      UMLS_syn_consolidated.txtNot used
      ecri.txtNot used
      ramiller-prune_kwf.txtNot used
    • iOnly Include those are Lexicon (add Eui).
    • Exclude those are already in synonymMeta, synonymNom, and synonymLvg
    • print to output file
  • Outputs:
    • ./outData/Results/synonymFromStmt.data.dup (exist sPairs)
    • ./outData/Results/synonymFromStmt.data
      NpLc Synonym-1Synonym-1Pos-1Synonym-2Pos-2STMT