A Python-based solution to detect and redact Personally Identifiable Information (PII) from CSV data streams, designed to prevent fraud incidents like those experienced by Flixkart.

• detector_dharaneesh_panneerselvam.py - Main PII detection and redaction script
• redacted_output_dharaneesh_panneerselvam.csv - Output file with detected and redacted PII
• deployment_strategy.md - Detailed deployment architecture proposal
• README.md - This documentation

python3 detector_dharaneesh_panneerselvam.py iscp_pii_dataset_-_Sheet1.csv

• Phone Numbers: 10-digit numbers
• Aadhar Cards: 12-digit numbers
• Passport Numbers: Format like P1234567
• UPI IDs: username@bank format (paytm, ybl, okaxis, etc.)

• Names: Full names or first+last name combinations
• Email Addresses: Valid email formats
• Physical Addresses: Street + city + pin code
• Device ID/IP Address: Only when combined with user context

• Single first/last names alone
• Standalone email addresses
• Individual city, state, or pin codes
• Transaction/order IDs
• Product descriptions

• Phone: 98XXXXXX10 (preserve first 2 and last 2 digits)
• Aadhar: 12XXXXXXXX12 (preserve first 2 and last 2 digits)
• Passport: PXXXXXX7 (preserve first and last character)
• UPI: usXXX@paytm (preserve first 2 chars and domain)
• Email: [email protected] (preserve first 2 chars and domain)
• Names: JXXX DXXX (preserve first character of each part)
• Address: Replace numbers with X's
• IP: Replace all numbers with XXX