Big Data-driven Approach for the Next Generation User Authentication

点击翻译按钮获取中文摘要

The remarkable growth in digital data is changing what and how the defense against the unknown will take place. Big data is a technical term used today to represent this massive growth of digital data that's being created from many sources. Organizations have turned their attentions to the deployment of Big Data analytics to gain valuable insights that benefit their businesses within protected and secure environments. Hence, network security protocols are being re-designed to manage the characteristics of Big Data which will eventually deliver the real benefits of this data growth. Contrary to the traditional perspective, in which researchers focus on identifying users' identity to protect Big Data-based environments, I have a contradictory perspective. I believe that Big Data itself will be the fuel for the next generation user authentication. In other words, the goal of this research program is to develop a new approach to information security that leverages Big Data analytics to provide true understanding of people's traits by which authentication decisions would be taken based on what users do on moment-to-moment basis. The proposed research program involves three short-term objectives that aim to investigate the development of the Big Data-driven authentication approach that in turn will provide new use cases for organizations who are looking for stronger authentication, cost-effective, and a convenient user experience. These three objectives will consider all aspects of the analysis, the design, and the implementation to deliver the following components. 1) Big Data analytics-based tools for measuring real-time human dynamics Investigate the deployment of large-scale data processing engines, such as Spark and Hadoop, to develop new Data Security-based Analytics (DSA) tool. Design methodology for building cognitive analytics methods to distinguish the data that has security/identification potentials. Finally, develop the Big Data-driven authentication model (BDA). 2) Big Data-driven Authentication as a Service Model Design and develop SaaS-based authentication (AUTHaaS ) model. A new approach for Big Data driven authentication that is powered by DSA, BDA, and JitHDA technologies initiated and shipped with the first component above. The services should be compliance with the data privacy legislation of Canada. 3) An integration framework to facilitate the collaboration and interoperability of multiple Big Data-driven authentication service providers Design and develop a new framework (iAUTH) to provide a unified interface necessary to leverage the DSA, BDA, and JitHDA technologies among multiple authentication service providers. This promotes simplicity and avoids any security chaos. The correctness of this framework will be verified using the formal method analysis “Syverson and Van Oorschot” SVO.