R²BN: An Adaptive Model for Keystroke-Dynamics-Based Educational Level Classification

Abstract

Over the past decade keystroke-based pattern recognition techniques as a forensic tool for behavioural biometrics have gained increasing attention. Although a number of machine learning based approaches have been proposed, they are limited in terms of their capability to recognise and profile a set of individual's characteristics. In addition, up to today their focus was primarily gender and age, which seem to be more appropriate for commercial applications (such as developing commercial software), leaving out from research other characteristics, such as the educational level. Educational level is an acquired user characteristic, which can improve targeted advertising, as well as provide valuable information in a digital forensic investigation, when it is known. In this context, this paper proposes a novel machine learning model, the Randomised Radial Basis function Network (R 2 BN), which recognises and profiles the educational level of an individual who stands behind the keyboard. The performance of the proposed model is evaluated by using empirical data obtained by recording volunteers' keystrokes during their daily usage of a computer. Its performance is also compared with other wellreferenced machine learning models using our keystroke dynamic datasets. Although the proposed model achieves high accuracy in educational level prediction of an unknown user, it suffers from high computational cost. For this reason, we examine ways to reduce the time that is needed to build our model, including the use of a novel data condensation method, and discuss the trade off between an accurate and a fast prediction. To the best of our knowledge, this is the first model in the literature that predicts the educational level of an individual based on keystroke dynamics information only.

Mobile keystroke dynamics for biometric recognition: An overview

“…Fifteen emotional states, including confidence, hesitance, nervousness, relaxation, sadness, and tiredness, have been estimated from typing patterns in [67]. Also the educational level of a subject has been predicted based on collected KD in [76]. Furthermore, several studies have also estimated soft biometrics such as gender, age, and handedness of subjects whose KD have been recorded through smartphones [77].…”

On the Inference of Soft Biometrics from Typing Patterns Collected in a Multi-device Environment

“…For example, Epp et al [Epp et al 2011] investigated the prediction of fifteen emotional states, including confidence, hesitance, nervousness, relaxation, sadness, and tiredness from typing patterns. Tsimperidis et al [Tsimperidis et al 2020] predicted the educational level of participants based on the keystroke dynamics information only. Beyond typing patterns, researchers have explored other behavioral patterns such as walking patterns, swiping patterns, calling patterns, device usage patterns to estimate a wide variety of soft identifiers [Acien et al 2018;Garofalo et al 2019;Miguel-Hurtado et al 2016b;.…”

Analysis of Doddington zoo classification for user dependent template update: Application to keystroke dynamics recognition

“…The adaptive strategies have been deeply used to enhance the performances of the biometric systems for different modalities [17] like face [18,19], voice [20] and keystroke dynamics [21]. Different types of adaptation process have been proposed in the literature:…”

“…• Adaptation of the system parameters: It generally consists in updating the parameters of the classifier depending on the user [22] or the quality of the capture [23]. Recently, in [21], the authors proposed an R 2 BN adaptive model that consists in increasing the weight on the misclassified instances to provide them to the next-level classifier to perform better. The authors 70 state that the proposed model achieved high accuracy in educational level prediction through the keystroke dynamics of the user.…”