Data anonymization is the process of removing sensitive information, such as names, addresses, personally identifiable information, and financial data from a text.
As well as ensuring confidentiality and compliance with regulations like GDPR, data anonymization increases content recycling, streamlines the enhancement process, and avoids unfortunate machine translations. The possibility of successful recycling data also increases significantly.
Here’s a summary of the anonymization steps we took when privacy by design at Exfluency was released:
- De-anonymized data is uploaded and stored on a secure server separate from the rest of the platform
- All confidential data is stored in client gated communities that can only be accessed by the Requester and those SMEs specifically cleared to work there
- Only anonymized data is sent to the four NMT engines
- All data sent to the NMT engines is first scrambled
- Only anonymized data is sent to the SMEs
- Only anonymized data is stored on the platform
- SMEs cannot copy-paste source data for extraction or use it in other tools
- Files are de-anonymized on the secure server before return to the Requester
- Project files on the secure server are destroyed
- Final versions are not available to anyone but the Requester