NLTK Gazetteers (Extended)

Geographic and demographic word lists, extended from the original NLTK gazetteers corpus.

Each config is one list; each row is one entry.

Original configs (NLTK gazetteers corpus)

Config	Description	License
countries	289 country names	GFDL (Wikipedia)
isocountries	234 ISO country names	public domain
nationalities	200 nationality adjectives	public domain
caprovinces	14 Canadian provinces	—
mexstates	32 Mexican states	—
usstates	52 US states	—
usstateabbrev	137 US state abbreviations	—
uscities	255 US cities 100k+ population	GFDL (Wikipedia)

Schema: name (string) — one entry per row.

Extended configs (new sources)

Config	Description	License
worldcities	33,602 cities ≥15k population	CC-BY 4.0 (GeoNames)
un-m49	249 countries with UN M49 regional codes	public domain
iso3166-2	6,260 country subdivisions (states, provinces, …)	Apache 2.0
cldr-{lang}	Territory names in 73 languages	Unicode License
wikidata-countries	310 countries with capital and demonym	CC0 (Wikidata)
osm-capitals	209 capital cities (OSM, ODbL)	ODbL

Schema by config

Config	Columns
worldcities	name, ascii_name, country_code, population, latitude, longitude, timezone
un-m49	name, alpha2, alpha3, m49_code, region, sub_region
iso3166-2	name, code, country_code, category
cldr-{lang}	code, name
wikidata-countries	name, capital, demonym
osm-capitals	name, country_code, latitude, longitude

All configs include name as the primary gazetteer string.

CLDR language codes

73 configs: cldr-af, cldr-ar, cldr-az, cldr-be, cldr-bg, cldr-bn, cldr-bs, cldr-ca, cldr-cs, cldr-cy, cldr-da, cldr-de, cldr-el, cldr-en, cldr-es, cldr-et, cldr-eu, cldr-fa, cldr-fi, cldr-fil, … and 53 more.

Usage

from datasets import load_dataset

# World cities (with country, population, coords)
ds = load_dataset("nltk-data-hub/gazetteers", "worldcities")
cities = ds["gazetteers"].to_pandas()

# Territory names in French
ds = load_dataset("nltk-data-hub/gazetteers", "cldr-fr")
territories = ds["gazetteers"]["name"]

# Countries with capitals and demonyms
ds = load_dataset("nltk-data-hub/gazetteers", "wikidata-countries")
df = ds["gazetteers"].to_pandas()

Via NLTK

import nltk
nltk.download("gazetteers")

nltk.corpus.gazetteers.words("countries.txt")
nltk.corpus.gazetteers.words("uscities.txt")

License

• Original configs: GFDL (Wikipedia-derived) / public domain
• worldcities: CC-BY 4.0 — GeoNames
• un-m49: public domain — UN Statistics Division
• iso3166-2: Apache 2.0 — ipregistry/iso3166
• cldr-*: Unicode License — CLDR
• wikidata-countries: CC0 — Wikidata
• osm-capitals: ODbL — OpenStreetMap