Abstract
In this article, we introduce a method of personal authentication with keystroke rhythm patterns based on Self-Organizing Maps (SOM). Our method is simple but convenient for login to Web application systems on the Internet as a low-cost biometric authentication, which is important in information security today. We implemented the prototype program in Java, and experimented with it. In the experiment, 10 subjects tried to type their own passphrase rhythmically in 4 times, and these collected rhythm patterns were evaluated with SOM. As a result, keystroke rhythm patterns are useful for personal authentication, so that both FAR (False Accept Rate) and FRR (False Reject Rate) are less than 8%. It is effective to use our method with other biometric authentication methods.
