avidas / node-genderize

Wrapper for the genderize API

genderize

Install with npm install genderize. This is a wrapper for the genderize web service.

Note that you should not use this to guess which gender you should assign to a person. Names are not a reliable predictor of which gender people want to use. I think it should only be used for analytical purposes of a set of names.

Example

var genderize = require('genderize')

genderize('Julia', function (err, obj) {
  console.log(obj.gender) // outputs 'female'
})

// optional localization parameters (see https://genderize.io/#localization)
genderize('Andrea', {language_id: 'it'}, function (err, obj) {
  console.log(obj.gender) // outputs 'male'
})

Stream

objectStream
  .pipe(genderize({language_id: 'it'}))
  .pipe(ndjson.stringify())
  .pipe(process.out)

List

genderize.list(['Julia', 'Finn', 'Christian', 'Andrea'], function (err, obj) {
  console.log(obj)
})

genderize.list(['Julia', 'Finn', 'Christian', 'Andrea'], {language_id: 'it'}, function (err, obj) {
  console.log(obj)
})

CLI

You can also do npm install genderize -g and you will get a simple streaming interface for piping in names and getting beck new line delimited JSON objects.

Given names.txt

"Julia"
Florian
Finn
Andrea

This command could result. Note that it doesn't have to be in the correct order.

$ genderize < names.txt
{"name":"Julia","gender":"female","probability":"0.99","count":2099}
{"name":"Florian","gender":"male","probability":"1.00","count":469}
{"name":"Finn","gender":"male","probability":"0.99","count":81}
{"name":"Andrea","gender":"female","probability":"0.79","count":5623}

$ genderize --language_id it < names.txt
{"name":"Julia","gender":"female","probability":"1.00","count":7,"language_id":"it"}
{"name":"Florian","gender":"male","probability":"1.00","count":4,"language_id":"it"}
{"name":"Finn","gender":null,"language_id":"it"}
{"name":"Andrea","gender":"male","probability":"0.98","count":1033,"language_id":"it"}