Sign In

Communications of the ACM

ACM TechNews

Exposing Flaws in Metrics for ­ser Login Systems

View as: Print Mobile App Share:
One of the most popular ways to log in involves typing user names and text-based passwords.

Researchers have found that the metrics commonly used in research to report the performance of user login systems are flawed, and pose a major security and privacy risk.

Credit: Can Liu/Rutgers University-New Brunswick

Researchers at Rutgers University–New Brunswick have found that metrics commonly used metrics in research to report the performance of user login systems are flawed, exposing a major security and privacy risk.

This means the systems may not work well, which can result in serious consequences for proposed systems that are adopted based on misleading metrics.

The researchers reviewed 35 recent research papers on authentication systems and found 94% of them had flaws in what they reported.

In addition, the team found that there is no consistent approach for reporting system performance metrics, and the metrics are inadequate.

The researchers developed a novel method that gives other academics, government agencies, and the public accurate information on the effectiveness of their authentication systems and how they can be improved.

The solution combines the strengths of a commonly used metric from other fields and a rarely used metric. When applied together, the metrics can be used to measure the success of user login systems.

From Rutgers Today
View Full Article


Abstracts Copyright © 2019 SmithBucklin, Washington, DC, USA


No entries found