Periodic analysis of database to maintain accuracy and integrity of AADHAAR

The methodology for generation of unique identity number (AADHAAR) for residents of India, involves usage of certain basic demographic (name, age, gender and address) and biometric information (ten finger prints, two iris images) with a photograph to uniquely identify a resident after a process of biometric de-duplication. The process of biometric de-duplication helps in ensuring uniqueness to a high degree of accuracy, exceeding 99%. The data collected during enrolment process is immediately encrypted and transmitted to Central Identities Data Repository (CIDR) of UIDAI for further processing. This information was given by the Minister of State (Independent Charge) for Planning, Shri Rao Inderjit Singh in a written reply in Rajya Sabha today.

The Minister said that after a series of validation/quality checks, the residents’ biometrics are matched against the existing database of UIDAI, on a 1:N matching basis, to determine if the resident has previously enrolled. Currently, over 700 trillion matches are carried out each day to ensure exhaustive de-duplication. The process involves various stages such as automated data validation, manual quality checks, demographic de-duplication, and biometric de-duplication. More than 9 crore enrollment packets have been rejected so far, that did not meet the quality and de-duplication criteria. Government is fully alive to the need of constantly monitoring and observing quality and veracity of Aadhaar database, and a dedicated Fraud Investigation and Analytics team has been deployed to ensure compliance with various fraud detection rules. Further, periodic exploratory data analysis of UIDAI database is undertaken to detect any fraudulent patterns in the system.

******

NCJ/NN