Big Data approach to biometric-based identity analytics

Abstract

Very large-scale biometric systems are becoming mainstream in nationwide identity cards and mobile secure payment methods. As with other Big Data systems, biometric systems contend with the 'four V' challenges that involve the effective managing of the complex life cycle and operations of identity information - despite the immense enrollment database size (volume) and rapid transaction response-time (velocity ) requirements using potentially noisy, fraudulent (veracity), and multiple (variety ) biometric identifiers. We describe representative techniques addressing these challenges in a variety of diverse and realistic biometrics applications such as 1) indexing methods that allow access in near constant time (velocity) irrespective of the scale (volume) of the biometric database, 2) fusing multiple (variety) biometrics to address risk and accuracy concerns, and 3) ensuring the integrity (veracity) of biometric databases by eliminating multiple duplicate records as well as protecting against theft of biometric identifiers. While sharing many commonalities with generic Big Data system-design issues, biometric systems also provide a rich case study involving how these issues manifest and are addressed in a unique, domain-specific way. We believe that by virtue of dealing with some of the most critical entities, namely identity and entitlement, biometric systems are likely to emerge as among the most critical of the Big Data systems.