Skip to content

anonymization_sample.md

Latest commit

 

History

History
89 lines (57 loc) · 3.95 KB

anonymization_sample.md

File metadata and controls

89 lines (57 loc) · 3.95 KB
Raw

Instructional Guide: Dataset Anonymization Process

This guide explains how to transform a raw dataset containing personally identifiable information (PII) into an anonymized version suitable for research and analysis.

1. Initial Data Collection

Initially, your dataset consists of identifiable files for 10 participants (e.g., Tom, Jerry, and 8 others):

  • 10 Past Transaction Logs: Tom_trans.xls, Jerry_trans.xls, etc.
  • 10 Signed Informed Consents: Tom_consent.pdf, Jerry_consent.pdf, etc.
  • 1 Demographics File: A master list containing participant info (age, zone, profession, etc.)

2. Pre-Anonymization View (Raw Data)

Sample: transactions.xls

Date Heure Direction Contact Téléphone Contenu Type
2025-02-21 09:30:07 Reçu OrangeMoney OrangeMoney Paiement de SEMBE WOILA en succès par 6995098203. ID transaction: MP250221.0930.C42749, Montant: 200 FCFA, Nouveau solde: 5534 FCFA SMS
2025-02-21 16:42:28 Reçu OrangeMoney OrangeMoney Paiement de SWITCHN réussi par 6995098203 JERRY THE MOUSE. ID transaction: MP250221.1642.C75125, Montant: 160 FCFA SMS
2025-02-22 07:46:51 Reçu OrangeMoney OrangeMoney Transfert de 6995098703 JERRY THE MOUSE vers 698016232 BLANCHE NEIGE réussi. ID transaction: MP250222.0746.C35694 SMS

Sample: Demographics (Raw)

User Age Range Gender Profession Education Monthly Income Geographic Zone
JERRY THE MOUSE 25–34 M Driver Bachelor 200,000–500,000 Urban (Yaoundé, Centre)
TOM THE CAT 18–24 M Student High School < 100,000 Suburban (Douala, Littoral)

3. The Anonymization Process

Step 1: Identification of Sensitive Information

We must locate all PII (Personally Identifiable Information) within the data:

  • Participant Names: TOM THE CAT, JERRY THE MOUSE
  • Third-Party Names: BLANCHE NEIGE, etc.
  • Phone Numbers: 6995098203, 696224574, etc.
  • Other Identifiers: Transaction IDs (which can be traced back to accounts)

Step 2: Anonymization Strategy

Apply the following replacement rules:

  1. Participant Names → User IDs

    • Example:
      • JERRY THE MOUSE → user0001
      • TOM THE CAT → user0002

  2. Other Names → "Mr. X"

  3. Phone Numbers → "XXXX"

  4. Transaction IDs → "ID_MASKED"

4. Anonymization Results (Final Data)

Updated File Structure

  • Transaction Logs: user0001.xls, user0002.xls, etc.
  • Demographics File: demographic_anonym.xls

Example: Content of user0001.xls

Date Heure Contact Contenu (anonymized)
2025-02-21 09:30:07 OrangeMoney Paiement de SWITCHN réussi par XXXX user0001. ID transaction: ID_MASKED, Montant: 150 FCFA
2025-02-22 07:46:51 OrangeMoney Transfert de XXXX user0001 vers XXXX Mr. X réussi. ID transaction: ID_MASKED, Montant Net: 1608 FCFA

Example: demographic_anonym.xls

User ID Age Range Gender Profession Geographic Zone
user0001 25–34 M Driver Urban (Yaoundé, Centre)
user0002 18–24 M Student Suburban (Douala, Littoral)