﻿# english
# admin region type prefix and suffix
# these are identified in location names and used for token expansion to include location type variants
# this gives us the maximum chance of matching a location name to the various ways to express it in text
# e.g. 'london st' -> 'london street', 'london st'

# TITLE = title word of placename (e.g. 'the')
# TYPE = type of place (e.g. 'road')
# +<phrase> = prefix
# <phrase>+ = suffix
# +<phrase>+ = can be both prefix and suffix
# *<phrase> = dont use this token for a match but do include it in token expansion
#   e.g. primary, *school --> matches only primary BUT will expand a match to <name> primary, <name> school
#        this allows avoidance of overmatching to general names like school (which can be primary, secondary etc.)

# note: save this as UTF-8 not a default latin encoded text file
# note: types are processed in reverse phrase gram order (so large phrases are matched first)

# add to stoplist location words that on thier own cannot be a specific location
# e.g. 'north' cannot be used for 'north london'
TITLE, +northeast+, +north east+, +north-east+
TITLE, +northwest+, +north west+, +north-west+
TITLE, +southeast+, +south east+, +south-east+
TITLE, +southwest+, +south west+, +south-west+
TITLE, +north+, +northern
TITLE, +south+, +southern
TITLE, +east+, +eastern
TITLE, +west+, +western
TITLE, +greater+
TITLE, +larger+
TITLE, +lesser+
TITLE, +smaller+
TITLE, +upper+
TITLE, +lower+
