Data Collection

Provision Methodology

The general workflow for data contribution is as follows:

1. The user requests data through the PrimeInsights Extension.

2. The PrimeInsights Extension fetches an Amazon Public Link from Amazon.com and retrieves the Amazon data (as a zip file).

3. The Amazon data is encrypted by the user.

4. The encrypted Amazon data is uploaded to IPFS.

5. The PrimeInsights server calculates the hash of the Amazon data.

6. The Amazon Public hash is stored on the PrimeInsights server.

7. The data registration includes the IPFS link and metadata.

8. A Docker Image is triggered to:

   1. Fetch the registered Amazon data hash and verify its integrity.

   2. Generate a proof to confirm data validity.

9. The proof and verification results are stored and returned to the user as a proof index.

10. IPFS nodes fetch and provide the encrypted Amazon data when requested.

Data Storage

PrimeInsights DLP employs IPFS (InterPlanetary File System) as the storage layer for ensuring decentralized and secure data handling. After user data is processed on the PrimeInsights website to remove PII, the sanitized data and associated proof are uploaded to an IPFS node. This ensures that both the original data's integrity and its authenticity proof are stored in an immutable, distributed manner, preventing unauthorized modifications. Upon request, a user can have his encryption keys burned by the DLP, ensuring that user data is unretrievable.

The IPFS node serves as a repository for the processed data and proof, providing a unique IPFS link that acts as a reference. These IPFS links are utilized throughout the system to validate and compute scores within the TEE Validator and Data Registry Contract. By leveraging IPFS, the system guarantees transparency, tamper-resistance, and accessibility, while maintaining user privacy.