ushahidi / geograpy Goto Github PK

View Code? Open in Web Editor NEW

219.0 24.0 135.0 1.28 MB

Extract countries, regions and cities from a URL or text

Python 100.00%

geograpy's Introduction

This project is no longer being maintained and has been archived. Please check the Forks list for newer versions.

Forks

We are aware of two 3rd party forks for this library:

[Maintained] https://github.com/somnathrakshit/geograpy3: recently revived, this project will be ensuring maintanance of Geograpy.Thanks to @WolfgangFahl for getting in touch about maintaining this.
[Outdated] This fork fixes issues with newer versions of nltk. A rewrite that fixes more issues is available here, please use it instead: https://github.com/Corollarium/geograpy2

Geograpy

Extract place names from a URL or text, and add context to those names -- for example distinguishing between a country, region or city.

Install & Setup

Grab the package using pip (this will take a few minutes)

pip install geograpy

Geograpy uses NLTK for entity recognition, so you'll also need to download the models we're using. Fortunately there's a command that'll take care of this for you.

geograpy-nltk

Basic Usage

Import the module, give some text or a URL, and presto.

import geograpy
url = 'http://www.bbc.com/news/world-europe-26919928'
places = geograpy.get_place_context(url=url)

Now you have access to information about all the places mentioned in the linked article.

places.countries contains a list of country names
places.regions contains a list of region names
places.cities contains a list of city names
places.other lists everything that wasn't clearly a country, region or city

Note that the other list might be useful for shorter texts, to pull out information like street names, points of interest, etc, but at the moment is a bit messy when scanning longer texts that contain possessive forms of proper nouns (like "Russian" instead of "Russia").

But Wait, There's More

In addition to listing the names of discovered places, you'll also get some information about the relationships between places.

places.country_regions regions broken down by country
places.country_cities cities broken down by country
places.address_strings city, region, country strings useful for geocoding

Last But Not Least

While a text might mention many places, it's probably focused on one or two, so Geograpy also breaks down countries, regions and cities by number of mentions.

places.country_mentions
places.region_mentions
places.city_mentions

Each of these returns a list of tuples. The first item in the tuple is the place name and the second item is the number of mentions. For example:

[('Russian Federation', 14), (u'Ukraine', 11), (u'Lithuania', 1)]

If You're Really Serious

You can of course use each of Geograpy's modules on their own. For example:

from geograpy import extraction

e = extraction.Extractor(url='http://www.bbc.com/news/world-europe-26919928')
e.find_entities()

# You can now access all of the places found by the Extractor
print e.places

Place context is handled in the places module. For example:

from geograpy import places

pc = places.PlaceContext(['Cleveland', 'Ohio', 'United States'])

pc.set_countries()
print pc.countries #['United States']

pc.set_regions()
print pc.regions #['Ohio']

pc.set_cities()
print pc.cities #['Cleveland']

print pc.address_strings #['Cleveland, Ohio, United States']

And of course all of the other information shown above (country_regions etc) is available after the corresponding set_ method is called.

Credits

Geograpy uses the following excellent libraries:

NLTK for entity recognition
newspaper for text extraction from HTML
jellyfish for fuzzy text match
pycountry for country/region lookups

Geograpy uses the following data sources:

GeoLite2 for city lookups
ISO3166ErrorDictionary for common country mispellings via Sara-Jayne Terp

Hat tip to Chris Albon for the name.

geograpy's People

Stargazers

Watchers

Forkers

gregdl corollarium yoopea wkryst spbriggs roosje spartakos87a mdjawad fil sharonanne potatochip davidcurtis llazzaro knusul zcarwile bclifton pelumi cordje codynhat cloudxtreme jhs2jhs sandy4321 kmanohar31 jay-chin surumen sokolandia benmaier l1kw1d harshil93r stranaeus iancamp newtonmwai xubeisi irr ingokl mikastefanie samzhang111 fdelorenzi bobquest33 lesingerouge w512 gaurangkher chiraggiri ubaid-me billytotochan pandawhocodes vgpprasad91 skvnet apahomov metavi jbritish skumaran141989 stasysp arju88nair suatatan labriffa charl4e shashankmehra jamesbsilva tavein avashek rohitjun08 barseghyanartur craig-matadeen kumarsanjeet1234 acarpentier madcap-lad moiex kprahman eug tixierae kirosg nihatha nawfalmn sergeigks woollard qianmao aniruthanv keineahnung2345 peshalaliyanage statsrick reginababo daoli cyolveren kyunbit parassharmaa tameem92 abhinavjain13 gdgcpb deeplogix-ashutosh bigapartmentsin jackxu-eb pratheepkanati mathieuchateau rt-hamilton seowon98 sierraa feras rizal-rovins asamadmcs7262

geograpy's Issues

in geograpy-nltk

the nltk.downloader() have been changed to nltk.download

geograpy-ntlk error

This is an issue fork from #4 by @shun-liang. I have the same problem.

When trying to run geograpy-nltk, I get the following error:

Traceback (most recent call last):
File "/Users/shun/.virtualenvs/hn_hiring_trend/bin/geograpy-nltk", line 5, in
nltk.downloader('maxent_ne_chunker')
TypeError: 'module' object is not callable

regions/countries returning all proper nouns

import geograpy as gp url = "https://www.politico.eu/article/italy-incurable-economy/" places = gp.get_place_context(url = url) places.regions

Returns a list of proper nouns from the article, the same goes for places.countries.

places.country_cities
seems to do better but still gives a funky return.
{'Italy': ['Rome', 'Naples', 'Codogno'], 'United States': ['Rome', 'Naples', 'Pierre', 'Brussels', 'Italy'], 'Belgium': ['Brussels'], 'France': ['Pierre']}

Does it only work with English?

Unable to run it due to label() exception on extraction.py

Traceback (most recent call last):
  File "ale.py", line 7, in <module>
    places = geograpy.get_place_context(url="https://www.cntraveler.com/hotels/hong-kong-s-a-r-/jordan/mandarin-oriental-hong-kong")
  File "/usr/local/lib/python2.7/site-packages/geograpy/__init__.py", line 6, in get_place_context
    e.find_entities()
  File "/usr/local/lib/python2.7/site-packages/geograpy/extraction.py", line 31, in find_entities
    if (ne.node == 'GPE' or ne.node == 'PERSON') and ne[0][1] == 'NNP':
  File "/usr/local/lib/python2.7/site-packages/nltk/tree.py", line 217, in _get_node
    raise NotImplementedError("Use label() to access a node label.")

All examples return empty result

import geograpy
from geograpy import places
from pprint import pprint

# address = geograpy.get_place_context(text='United States')
# pprint (vars(address))
from geograpy import extraction

e = extraction.Extractor(url='http://www.bbc.com/news/world-europe-26919928')
e.find_entities()

# You can now access all of the places found by the Extractor
pprint (e.places)
pprint (vars(e.places))

Results:

{'address_strings': [],
 'cities': [],
 'city_mentions': [],
 'conn': <sqlite3.Connection object at 0x110769858>,
 'countries': [],
 'country_cities': {},
 'country_mentions': [],
 'country_regions': {},
 'other': [],
 'places': [],
 'region_mentions': [],
 'regions': []}
[]

UnicodeDecodeError: 'charmap' codec can't decode...

Did @PandaWhoCodes pip install git+https://github.com/reach2ashish/geograpy.git plus
nltk.downloader.download('maxent_ne_chunker')
nltk.downloader.download('words')
nltk.downloader.download('treebank')
nltk.downloader.download('maxent_treebank_pos_tagger')
nltk.downloader.download('punkt')
nltk.download('averaged_perceptron_tagger')

and it seemed to be going well until I tried the example
url = 'http://www.bbc.com/news/world-europe-26919928'
places = geograpy.get_place_context(url=url)

I get
UnicodeDecodeError: 'charmap' codec can't decode byte 0x8d in position 274: character maps to <undefined>

Python 3.6 Windows
Any thoughts? (or alternatives? I need to pull out city names. I've used GeoText for the country names (not positive it's working right yet) but GeoText's cities doesn't work very well.)

sqlite3.OperationalError

Fresh install on Python3 on Ubuntu 16.04

import geograpy
address = geograpy.get_place_context(text='United States')

results in

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/local/lib/python3.5/dist-packages/geograpy/__init__.py", line 9, in get_place_context
    pc = PlaceContext(e.places)
  File "/usr/local/lib/python3.5/dist-packages/geograpy/places.py", line 17, in __init__
    self.conn = sqlite3.connect(db_file)
sqlite3.OperationalError: unable to open database file

ImportError: cannot import name 'PlaceContext'

ImportError Traceback (most recent call last)
in ()
----> 1 from geograpy import places

~\Anaconda3\envs\goodreads\lib\site-packages\geograpy-0.3.7-py3.6.egg\geograpy_init_.py in ()
1 from extraction import Extractor
----> 2 from places import PlaceContext
3
4 def get_place_context(url=None, text=None):
5 e = Extractor(url=url, text=text)

ImportError: cannot import name 'PlaceContext'

Can Someone Help me please?

Let people know about never versions and stackoverflow questions

https://stackoverflow.com/questions/tagged/geograpy has now a list of questions about geograpy and it's different versions:
See https://stackoverflow.com/tags/geograpy/info

https://github.com/somnathrakshit/geograpy3 has been revived today and has a python3 compatible version that has been tested with python 3.6, 3.7 and 3.8
Please add newer issues there so that they can be propery fixed.

NotImplementedError: Use label() to access a node label.

pip installed geograpy on archlinux box and tried out this simple program

import geoprapy
t = 'India is a country'
p = geoprapy.get_place_context(text=t)

I got the following error

In [3]: p = geograpy.get_place_context(text=txt)
---------------------------------------------------------------------------
NotImplementedError                       Traceback (most recent call last)
<ipython-input-3-ae088bc75e75> in <module>()
----> 1 p = geograpy.get_place_context(text=txt)

/home/m/.local/lib/python2.7/site-packages/geograpy/__init__.pyc in get_place_context(url, text)
      4 def get_place_context(url=None, text=None):
      5     e = Extractor(url=url, text=text)
----> 6     e.find_entities()
      7
      8     pc = PlaceContext(e.places)

/home/m/.local/lib/python2.7/site-packages/geograpy/extraction.pyc in find_entities(self)
     29         for ne in nes:
     30             if len(ne) == 1:
---> 31                 if (ne.node == 'GPE' or ne.node == 'PERSON') and ne[0][1] == 'NNP':
     32                     self.places.append(ne[0][0])

/usr/lib/python2.7/site-packages/nltk/tree.pyc in _get_node(self)
    196     def _get_node(self):
    197         """Outdated method to access the node value; use the label() method instead."""
--> 198         raise NotImplementedError("Use label() to access a node label.")
    199     def _set_node(self, value):
    200         """Outdated method to set the node value; use the set_label() method instead."""

NotImplementedError: Use label() to access a node label.

Python version 2.7.11
Linux zero 4.5.4-1-ARCH #1 SMP PREEMPT Wed May 11 22:21:28 CEST 2016 x86_64 GNU/Linux

ImportError: No module named 'extraction'

Hi,
I just came across this library and was trying it out but every time I try to....

import geograpy

I get this error:
Traceback (most recent call last):
File "", line 1, in
File "/Users/Larry/miniconda3/envs/fido/lib/python3.5/site-packages/geograpy/init.py", line 1, in
from extraction import Extractor
ImportError: No module named 'extraction'

Error processing data (from demo)

NLTK seems to have changed this: http://www.nltk.org/_modules/nltk/tree.html

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/local/lib/python2.7/dist-packages/geograpy/**init**.py", line 6, in get_place_context
    e.find_entities()
  File "/usr/local/lib/python2.7/dist-packages/geograpy/extraction.py", line 31, in find_entities
    if (ne.node == 'GPE' or ne.node == 'PERSON') and ne[0][1] == 'NNP':
  File "/usr/local/lib/python2.7/dist-packages/nltk/tree.py", line 198, in _get_node
    raise NotImplementedError("Use label() to access a node label.")
NotImplementedError: Use label() to access a node label.

geograpy wont download

I get this error when I run the command pip install geograpy then other errors return also.
"Exception:
Traceback (most recent call last):
File "/Library/Python/2.7/site-packages/pip-9.0.1-py2.7.egg/pip/basecommand.py", line 215, in main
status = self.run(options, args)"

Way to Reset the initial Place Context

Hi,

Ideally I would like if we could reset the place context. It is not being reset. So, let' say I initialize it with Afghanistan and then on next iteration add Libya. It will still display Afghanistan in country_mentions. It should be reset.

python 3 support

the module newspaper won't download because it has changed to newspaper3k. The library itself also seems to have some python3 incompatibilities that could be fixed by running 2to3.

Error in installation

I am getting following error while installation.

Could not find a version that satisfies the requirement geograpy (from versions: )
No matching distribution found for geograpy.

OperationalError: unable to open database file

plz I am having a big error here
Traceback (most recent call last):
File "sm.py", line 36, in
pc = places.PlaceContext(['Cleveland', 'Ohio', 'United States'])
File "/usr/local/lib/python3.6/dist-packages/geograpy3/places.py", line 34, in init
self.conn = sqlite3.connect(db_file)
sqlite3.OperationalError: unable to open database file

AttributeError: 'NoneType' object has no attribute 'name'

p = geograpy.get_place_context(text='Pristina')

Traceback (most recent call last):
File "", line 1, in
File "/home/cusco/VirtualEnvs/data_parser/lib/python3.7/site-packages/geograpy/init.py", line 12, in get_place_context
pc.set_cities()
File "/home/cusco/VirtualEnvs/data_parser/lib/python3.7/site-packages/geograpy/places.py", line 160, in set_cities
country_name = country.name
AttributeError: 'NoneType' object has no attribute 'name'

'NoneType' object has no attribute 'name'