Skip to content

Latest commit

 

History

History
64 lines (37 loc) · 2.12 KB

section_phonetic_correction.md

File metadata and controls

64 lines (37 loc) · 2.12 KB

Phonetic correction

Meyer, Mayer, Maier?

Notes:


Soundex

  • ­ For proper names, brand names, drugs
  • ­ Encode strings according to their sounds
  • ­ MeierM600

Notes:


Index

meier

M600

Term (not in index) Soundex Doc IDs
meier, maier, mayer, meyer M600 #1, #3, #5
müller, mueller M460 #2, #4, #6
goethe, göthe G300 #7, #8, #9

#1, #3, #5

Notes:


Precision and recall?

  • ­ Recall up
  • ­ Precision down

Notes:

  • How does this affect precision and recall?

Soundex alternatives

  • Daitch–Mokotoff Soundex
  • Metaphone
  • Double Metaphone

Notes: