Skip to main content

Genderizer tries to infer gender information looking at first name and/or making text analysis

Project description


Genderizer is a language independent module which tries to detect gender by looking given first names and/or analyzing sample texts.

You can install this package using the following ```pip``` command:

$ sudo pip install genderizer


from genderizer.genderizer import Genderizer

print Genderizer.detect(firstName = 'John')
# >>> male

print Genderizer.detect(firstName = 'Marry')
# >>> female

print Genderizer.detect(firstName = 'mustafa')
# >>> male

print Genderizer.detect(firstName = 'fatma')
# >>> female

print Genderizer.detect(firstName = 'fikret')
# >>> None

print Genderizer.detect(firstName = 'fikret', text='galatasary maçı kaçmaz')
# >>> male

print Genderizer.detect(firstName = 'fikret', text='annemlerle yemek yedik')
# >>> female

print Genderizer.detect(text='askerlik yoklamasını kaçırdım mk')
# >>> male

print Genderizer.detect(text='bana çiçek alan erkek için canım feda')
# >>> female

print Genderizer.detect(firstName='fatma', text='askerlik yoklamasını kaçırdım mk')
# >>> female

print Genderizer.detect(text='futbol sevgi')
# >>> None

print Genderizer.detect(text='lan bi siktir git')
# >>> male

***Note***: You may claim that women can say ```lan bi siktir git```, of course. But the probability of being female is less than the probability of being male according the trained data of the classifier.

By the way, in Turkish saying 'lan bi siktir git' makes you quite rude.

## How It Works
Genderizer is a module which tries to detect gender by looking given first names and/or analyzing sample text of a person.

If a first name is definitely used for only one gender, the system will accept this gender and will not make any further analysis. For example, while 'Mustafa', 'Osman', 'Hasan' is used in Turkish only for male; 'Fatma', 'Ceyda', 'Elif' only for female.

When looking at first names does not infer any gender for sure, the system will make text analysis if it is given. For example; 'Ekim', 'Meric', or 'Tümay' is used for both male and female.

The text analysis is the classification of sample texts. It simply try to compute the probability of being male or female mining the sample text. In this system Naive Bayesian Classification is adopted and [naiveBayesClassifier][1] is used.

## How To Improve It
TODO: write a step by step guide

##Preparing Language Dependent Training Sets
TODO: give a few examples

## Customization and Optimization
TODO: Tell what options are there and when to choose
from genderizer.memcachedNamesCollection import MemcachedNamesCollection
from genderizer.mongoNamesCollection import MongoNamesCollection

MongoNamesCollection.mongodbURL = 'mongodb://'

#For memcached, you need to setup a memcached server.
MemcachedNamesCollection.memcacheHost = ''

classifier=Classifier(CachedModel.get('tr'), tokenizer)

print Genderizer.detect(firstName = 'fikret', text='annem')


* inline docs
* unit-tests

* Mustafa Atik [@muatik][2]
* Nejdet Yucesoy @nejdetckenobi


Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Filename, size & hash SHA256 hash help File type Python version Upload date
genderizer- (9.4 MB) Copy SHA256 hash SHA256 Source None Apr 17, 2014

Supported by

Elastic Elastic Search Pingdom Pingdom Monitoring Google Google BigQuery Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page