Authentication and Identification Methods Used in Keystroke Biometric Systems

Abstract

Keystroke dynamics authentication is not as widely used compared to other biometric systems. In recent years, keystroke dynamic authentication systems have gained interest because of low cost and integration with existing security systems. Many different methods have been proposed for data collection, feature representation, classification, and performance evaluation. The work presents a detailed survey of the most recent research in keystroke dynamic authentication. Research is evaluated by the conditions under which data was collected, classification algorithms used, and system performance. This work also identifies some shortcomings of the current research issues that need to be addressed for keystroke dynamics to mature. Some recommendations for future research are made, with the goal of improving keystroke dynamics system performance and robustness.

A novel security scheme for behavioral authentication systems based on keystroke dynamics

“…Typing behavior can be extracted using existing hardware such as the regular standard keyboard on desktop personal computers and laptops as long as there are virtual keyboards in the touch screen devices. This makes this scheme inexpensive and extremely attractive technique …”

“…In NN models, researchers first build a prediction model from input provided by users during enrollment, which will be later used to predict the outcome of a new trial provided by the user during authentication. Authentication in NN, similar to statistical approach, consists of comparing the users profile by the new input provided in the authentication process , Hybrid techniques…”

“…This makes this scheme inexpensive and extremely attractive technique. 19,20 Behavioral rhythm is being characterized using several combinations of features. 21,22 These features are being extracted and processed from authentication system interfaces.…”

“…Authentication in NN, similar to statistical approach, consists of comparing the users profile by the new input provided in the authentication process. 7,8,[19][20][21][22][23][24][25] 3 Hybrid techniques Others techniques are based on combing both statistical and NNs techniques. These are often called hybrid techniques where hybrids methods are used to analyze the KSD features.…”

Enhanced Authentication System Performance Based on Keystroke Dynamics using Classification algorithms

“…A number of studies have been performed in the area of keystrokes analysis since it is founded on the field of biometric based systems [10][11][12][13][14][15]. Many authors proposing good contribution trials for existing unique datasets benchmarks [14] [15]. It was difficult to compare the performance among different researches based on KSD, so different public datasets were published in this field to provide one unique evaluation research based on one reference.…”