Combined scaled manhattan distance and mean of horner’s rules for keystroke dynamic authentication

Abstract

Account security was determined by how well the security techniques applied by the system were used. There had been many security methods that guaranteed the security of their accounts, one of which was Keystroke Dynamic Authentication. Keystroke Dynamic Authentication was an authentication technique that utilized the typing habits of a person as a security measurement tool for the user account. From several research, the average use in the Keystroke Dynamic Authentication classification is not suitable, because a user's typing speed will change over time, maybe faster or slower depending on certain conditions. So, in this research, we proposed a combination of the Scaled Manhattan Distance method and the Mean of Horner's Rules as a classification method between the user and attacker against the Keystroke Dynamic Authentication. The reason for using Mean of Horner's Rules can adapt to changes in values over time and based on the results can improve the accuracy of the previous method.

A Systematic Literature Review on Latest Keystroke Dynamics Based Models

“…These metrics are common in the user authentication model, which is widely accepted in industry and academia [327]. Accuracy is also a metric used in several studies [26], [195], [231], [291], [297], [328]. A study [241] used three metrics -FAR, FRR, and EER to measure the performance of the support vector-based model.…”

Continuous User Authentication Featuring Keystroke Dynamics Based on Robust Recurrent Confidence Model and Ensemble Learning Approach

“…Other statistical methods used for classification of keystroke dynamics in literature includes Euclidean distance [24], scaled Euclidean distance [25], scaled Manhattan distance with Mean of Horners Rules [26], Mahalanobis distance [21], and Bhattacharyya distance with Gaussian mixture model [27]. In addition, other statistical techniques such as Hidden Markov Model [28], Kolmogorov-Smirnov Test (KS-test) [29] , and Bayesian Classification [30], were also employed to find the level of similarity between keystroke samples.…”