Multiword Candidates Generation
Expansions of acronyms Pattern in the Medline N-gram Set

• The expansion of acronyms from the latest Medline N-gram Set is a good source for multiword candidates.
  • Many expansion of acronyms follow certain pattern in the Medline. The pattern is like: acronym expansion (ACRONYM)
  • Examples:

    Example	Notes
    zona pellucida (ZP)	E0216465
    major hydrophilic region (MHR)	E0760361
    diabetic foot syndrome (DFS)	E0564279
    diabetic foot ulcer (DFU)	E0715662
    major hydrophilic region (MHR)	E0760361
    years lived with disability (YLD)	Invalid LMWs No specific meanionmg no POS
    persons who stutter (PWS)	Invalid LMWs No POS No morphology rules
    violence against women (VAM)	Invalid LMWs No POS No morpho logy rules
    zero-point energy (ZPE)	Invalid LMWs low frequency

  Typically, the release of Medline N-gram set (MNS) is 3~9 months behind the release of the Lexicon.

• Matcher (ACR): Steps 1-3 (pre_process), Steps 4-5 (Process)
• Tag [y|n] for [valid|invalid] LMWs
  • [y]: include base, spVar, inflectional variant, ignore case
    => Add to Lexicon
  • No need to tag valid expansion
• Use 00.CandidateList to:
  • Remove candidates in the Lexicon:
  • Tag invalid LMWs (AUTO_N) based on previous tagging.
  • Monitor AUTO_N
• Generated files:
  Dir: ${MULTIWORDS}/data/${YEAR}/outData/07.MatcherParAcr/1.GenLmwCandidates/CandList/acronymExp.tag.data.tag.final.tbd.${YEAR}.used.rmYesTagNo

  Distilled MEDLINE nGram Set	Candidate Files	Status	Notes
  2015	acronymExp.tag.data.tag.final.tbd.2015	Done	Tag [y|n]
  2016	acronymExp.tag.data.tag.final.tbd.2016	Done	Tag [y|n]
  2017	acronymExp.tag.data.tag.final.tbd.2017	Done	Tag [y|n]
  2018	acronymExp.tag.data.tag.final.tbd.2018	Done	Tag [y|n] Monitor AUTO_N
  2019	acronymExp.tag.data.tag.final.tbd.2019.Used.rmYesNo	Done	No Tag Auto monitor AUTO_N
  2020	acronymExp.tag.data.tag.final.tbd.2020.used.rmYesNo	Done	No Tag Auto monitor AUTO_N
  2021	acronymExp.tag.data.tag.final.tbd.2021.used.rmYesNo	Done	No Tag Auto monitor AUTO_N