Authentication via keystroke dynamics

Keystroke dynamics as a biometric for authentication

“…It has already been shown (see [14,18,20]) that keystroke rhythm is a good sign of identity. Moreover, unlike other biometric systems which may be expensive to implement, keystroke dynamics is almost free -the only hardware required is the keyboard.…”

“…The rationale for this approach is that digraphs with abnormally long latencies are not likely to be representative of the authorized user's typing. While this seems like a reasonable assumption it has recently been shown ( [18,20]) that one filter value for all typists does not yield optimal performance.…”

“…Therefore, the 500 ms low-pass filter used by [17] excludes many keystrokes typical of novice typists, while at the same time, includes many keystrokes which are not representative of an expert typist [18]. Studies by [18,20] showed that the use of digraph-specific measures of variability instead of one low-pass filter can lead to measurable improvements in verification accuracy. Moreover, the approach of [17] to keystroke verification uses the key down-to-down time as the base unit of measure, but this measure may be further delineated into two orthogonal components -total time the first key is depressed (i.e.…”

“…Previous works [3,18,20] used these two components in their verification systems. However, the initial sample sets of [3,20] did not provide enough data to ascertain whether the use of the two separate orthogonal digraph components added significant predictive power to the more traditional key down-to-down measure. Substantially improved performance results based on using the bivariate measure of latency with an appropriate distance measure were achieved by [18].…”

“…• let x i be the feature vector, σ i the interclass dispersion vector and w i the weight vector, then the dis- 1 The results reported here reflect a larger sample set than that use in [20]. tance of two feature vectors x i and x i is expressed as:…”