Keystroke Dynamics as a Language Profiling Tool: Identifying Mother Tongue of Unknown Internet Users

Abstract

Understanding the distinct characteristics of unidentified Internet users is helpful in various contexts, including digital forensics, targeted advertising, and user interaction with services and systems. Keystroke dynamics (KD) enables the analysis of data derived from a user’s typing behaviour on a keyboard as one approach to obtain such information. This study conducted experiments on a developed dataset that recorded samples of typing in five different mother tongues to determine Internet users’ mother tongue. Based on only a few KD features and machine learning techniques, 82% accuracy was achieved in recognising an unknown user’s mother tongue. This research highlights the potential for KD as a reliable method for identifying the mother tongue of Internet users, with implications for various applications such as improving digital forensic investigations, targeted advertising strategies, and optimising user experiences with online services.

The Way We Type Reveals Our Native Language

“…The extensions of the methodology in relation to that of the paper [3] are found, firstly to the use of more keystroke dynamics features, secondly to the use of feature selection algorithm, and thirdly to the use of meta-algorithms to improve the performance of the system.…”

“…An important difference of the present work, which as mentioned above is an extension of [3], is that the data were collected with a free text method, which better approximates the conditions of real typing. From the data collected, in [3] keystroke durations were used for native language classification.…”

“…An important difference of the present work, which as mentioned above is an extension of [3], is that the data were collected with a free text method, which better approximates the conditions of real typing. From the data collected, in [3] keystroke durations were used for native language classification. The differences of the methodology of this work are that firstly, many more keystroke dynamics features were extracted, secondly a feature selection algorithm was used to use only the most suitable features and reduce the time required to extract the result, and thirdly a boosting algorithm was used to improve system performance.…”

“…The main goal of this paper is to develop a new approach to determine the native language of a user through keystroke dynamics, by using free-text data, in order to reveal part of the identity of an unknown person. This is achieved by extending the research presented in [3]. The proposed approach is based on the idea that different native languages have a unique effect on the way people type.…”

“…Almost all of user classification works were aimed at finding the gender and/or age of a user. Very few studies focus on user characteristics other than gender and age, and only three, [3], [19], and [20], deal with users' native language. In the latter two of these studies, the fixed text method was followed, in which volunteers are asked to type specific text.…”