tommy/libpostal
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[dictionaries] Splitting English dictionaries into distinct level types (numbered, basement, mezzanine, etc.) for adding tokens automatically in address parser training data

Browse Source
This commit is contained in:
Al
2016-03-17 13:05:55 -04:00
parent f1334c5111
commit 40cf3d1933
4 changed files with 20 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
resources/dictionaries/en/level_types_basement.txt Normal file
View File

@@ -0,0 +1 @@
basement|bsm|bsmt|bsmnt|basement|bsment

1
resources/dictionaries/en/level_types_mezzanine.txt Normal file
View File

@@ -0,0 +1 @@
mezzanine|mezz

6
resources/dictionaries/en/level_types_numbered.txt Normal file
View File

@@ -0,0 +1,6 @@
floor|fl|flr|/ f
level|lev|levl|lvel|lvl|l
platform|pf
podium|pd
rooftop|rt|rf
upper|upr

12
resources/dictionaries/en/level_types_standalone.txt Normal file
View File

@@ -0,0 +1,12 @@
ground|g|gd
ground floor|g|gd|gdfl|gd fl|gd/fl|gd / fl|gf|g / f
lobby
lower ground floor|lg|lgf|lgfl|l / g|l / gf|l / g / f|l / g / fl
lower level|lwr level|lower lvl|lwr lvl
lower left|lower l|lwr l
lower right|lower r|lwr r
rooftop|rt|rf|r / t
upper ground floor|ug|ugf|ugfl|ug / f|ug / fl
upper|uppr|upr
upper left|upper l|upr left|uppr left|upr l|uppr l
upper right|upper r|upr right|uppr right|upr r|uppr r