Skip to content

Text anonymization in many languages using Faker

License

Notifications You must be signed in to change notification settings

hal9ai/anonymization

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

17 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Anonymization

Text anonymization in many languages for python3.6+ using Faker.

Install

pip install anonymization

Example

Replace emails and named entities in english

This example use NamedEntitiesAnonymizer which require spacy and a spacy model.

pip install spacy
python -m spacy download en_core_web_lg
>>> from anonymization import Anonymization, AnonymizerChain, EmailAnonymizer, NamedEntitiesAnonymizer

>>> text = "Hi John,\nthanks for you for subscribing to Superprogram, feel free to ask me any question at [email protected] \n Superprogram the best program!"
>>> anon = AnonymizerChain(Anonymization('en_US'))
>>> anon.add_anonymizers(EmailAnonymizer, NamedEntitiesAnonymizer('en_core_web_lg'))
>>> anon.anonymize(text)
'Hi Holly,\nthanks for you for subscribing to Ariel, feel free to ask me any question at [email protected] \n Ariel the best program!'

Or make it reversible with pseudonymize:

>>> from anonymization import Anonymization, AnonymizerChain, EmailAnonymizer, NamedEntitiesAnonymizer

>>> text = "Hi John,\nthanks for you for subscribing to Superprogram, feel free to ask me any question at [email protected] \n Superprogram the best program!"
>>> anon = AnonymizerChain(Anonymization('en_US'))
>>> anon.add_anonymizers(EmailAnonymizer, NamedEntitiesAnonymizer('en_core_web_lg'))
>>> clean_text, patch = anon.pseudonymize(text)

>>> print(clean_text)
'Christopher, \nthanks for you for subscribing to Audrey, feel free to ask me any question at [email protected] \n Audrey the best program!'

revert_text = anon.revert(clean_text, patch)

>>> print(text == revert_text)
true

Replace a french phone number with a fake one

Our solution supports many languages along with their specific information formats.

For example, we can generate a french phone number:

>>> from anonymization import Anonymization, PhoneNumberAnonymizer
>>>
>>> text = "C'est bien le 0611223344 ton numéro ?"
>>> anon = Anonymization('fr_FR')
>>> phoneAnonymizer = PhoneNumberAnonymizer(anon)
>>> phoneAnonymizer.anonymize(text)
"C'est bien le 0144939332 ton numéro ?"

More examples in /examples

Included anonymizers

Files

name lang
FilePathAnonymizer -

Internet

name lang
EmailAnonymizer -
UriAnonymizer -
MacAddressAnonymizer -
Ipv4Anonymizer -
Ipv6Anonymizer -

Phone numbers

name lang
PhoneNumberAnonymizer 47+
msisdnAnonymizer 47+

Date

name lang
DateAnonymizer -

Other

name lang
NamedEntitiesAnonymizer 7+
DictionaryAnonymizer -
SignatureAnonymizer 7+

Custom anonymizers

Custom anonymizers can be easily created to fit your needs:

class CustomAnonymizer():
    def __init__(self, anonymization: Anonymization):
        self.anonymization = anonymization

    def anonymize(self, text: str) -> str:
        return modified_text
        # or replace by regex patterns in text using a faker provider
        return self.anonymization.regex_anonymizer(text, pattern, provider)
        # or replace all occurences using a faker provider
        return self.anonymization.replace_all(text, matchs, provider)

You may also add new faker provider with the helper Anonymization.add_provider(FakerProvider) or access the faker instance directly Anonymization.faker.

Benchmark

This module is benchmarked on synth_dataset from presidio-research and returns accuracy result(0.79) better than Microsoft's solution(0.75)

You can run the benchmark using docker:

docker build . -f ./benchmark/dockerfile -t anonbench
docker run -it --rm --name anonbench anonbench

License

MIT

About

Text anonymization in many languages using Faker

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%