Menu

Netanos

Named entity-based Text Anonymization for Open Science

Remove identifiable information from text files.
Web-based Netanos
Open-source for open science.

About Netanos

What is Netanos?

Netanos stands for Named entity-based Text Anonymization for Open Science.
We hope this tool helps promoting open science practices by allowing researchers who work with text data to share their data even if these may contain identifiable information. Netanos is documented in a software paper and has been validated experimentally with human participants

For whom is it?

This website provides a simple interface of the original Netanos software. It will be particularly useful for researchers who work with text data (e.g., user statements, transcripts, official documents) and want to remove identifiable information. Everyone is free to use Netanos.

Why?

Text data are rarely shared if they contain some potentially identifiable information (e.g., names, locations). This is unfortunate and hampers open science. Since its release in Summer 2017, we have worked on Netanos to provide this simple drag-and-drop interface. If you are interested in the developer version, check out GitHub and npm.

How does it work?

In essence, Netanos anonymizes (bunches of) text files through a technique called named entity recognition. This will allow you to share text files that contain identifiable information (e.g., names, locations, numbers) without violating privacy agreements. A detailed list of features can be found here (p. 10)

What happens to my text data?

Nowhere in the anonymization process are any data stored on servers. Once you input your text files, you will retrieve the anonymized files for a download. We are not able to access any of the text files during or after the anonymization.

Who's behind Netanos

The software was developed at the University of Amsterdam and is maintained by Bennett Kleinberg, Maximilian Mozes.