-
-
We look forward to taking this further with the sponsors of the Legaltech workshop!
The app recognizes context and does not just delete it but replace it by references to broader demographic categories. This makes the anonymized text compliant with anonymity requirements while still allowing the reader to understand and relate. This again increases user productivity.
The video demo has examples. For example broader geographic references are more useful than just redactions.
The app comes with an intuitive user interface and a case history, allowing the user to review all choices made by the app.
Herein lies also the differentiation to products currently on the market. This app will continue to learn.
The app will be open-sourced, and in return users will review anonymizations and rate or improve them.
Further work: understand context even further. The video shows an example of someone who bought then resold a car with distinctive mileage. This can ultimately also be an identifier, taken in context. The app will understand, improve an ultimately replace with broader references 'A car of more than 5 years with considerable mileage'.
Log in or sign up for Devpost to join the conversation.