CLICKA: Collecting and leveraging identity cues with keystroke dynamics

Buckley, Oliver; Hodges, Duncan; Windle, Jonathan; Earl, Sally

doi:10.1016/j.cose.2022.102780

Cited by 4 publications

(5 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is because the analysis of the collected data revealed that users who natively use various keyboard layouts have distinct ways of entering special characters or capital letters. For CLICKA researchers, attempts to recognize the native language of users based on keystroke dynamics allowed them to achieve only slightly higher accuracy than guessing [70]. With keylogging and mapping keys to appropriate behavior, the accuracy of such research would increase significantly.…”

Section: Discussionmentioning

confidence: 99%

A Deep-Learning-Based Approach to Keystroke-Injection Payload Generation

Gurčinas,

Dautartas,

Janulevičius

et al. 2023

Electronics

View full text Add to dashboard Cite

Investigation and detection of cybercrimes has been in the spotlight of cybersecurity research for as long as the topic has existed. Modern methods are required to keep up with the pace of the technology and toolset used to facilitate these crimes. Keystroke-injection attacks have been an issue due to the limitations of hardware and software up until recently. This paper presents comprehensive research on keystroke-injection payload generation that proposes the use of deep learning to bypass the security of keystroke-based authentication systems focusing on both fixed-text and free-text scenarios. In addition, it specifies the potential risks associated with keystroke-injection attacks. To ensure the legitimacy of the investigation, a model is proposed and implemented within this context. The results of the implemented implant model inside the keyboard indicate that deep learning can significantly improve the accuracy of keystroke dynamics recognition as well as help to generate effective payload from a locally collected dataset. The results demonstrate favorable accuracy rates, with reported performance of 93–96% for fixed-text scenarios and 75–92% for free-text. Accuracy across different text scenarios was achieved using a small dataset collected with the proposed implant model. This dataset enabled the generation of synthetic keystrokes directly within a low-computation-power device. This approach offers efficient and almost real-time keystroke replication. The results obtained show that the proposed model is sufficient not only to bypass the fixed-text keystroke dynamics system, but also to remotely control the victim’s device at the appropriate time. However, such a method poses high security risks when deploying adaptive keystroke injection with impersonated payload in real-world scenarios.

show abstract

Section: Discussionmentioning

confidence: 99%

A Deep-Learning-Based Approach to Keystroke-Injection Payload Generation

Gurčinas,

Dautartas,

Janulevičius

et al. 2023

Electronics

View full text Add to dashboard Cite

show abstract

“…In [19] the familiarity that a user acquires when typing the digrams contained in his/her name was exploited, while the data were collected with a fixed text method. In a similar way, the data were collected in [20], in which the researchers studied the way users type words in languages they are familiar with, as well as in those they are not.…”

Section: A Contribution Of Workmentioning

confidence: 99%

“…In their paper, Buckley et al [19] build upon the notion of keystroke dynamics to infer an anonymous user's name and predict their native language. For this purpose, they asked volunteers to type a paragraph, generated according to their names, three times.…”

Section: Introductionmentioning

confidence: 99%

“…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.…”

Section: Introductionmentioning

confidence: 99%

“…In the latter two of these studies, the fixed text method was followed, in which volunteers are asked to type specific text. In [19] a percentage of 45% is achieved in the correct prediction of the user's native language, among five different options, while in [20] some conclusions are drawn about the typing habits of users, related to their native language, so that they can be used in authentication systems.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

The Way We Type Reveals Our Native Language

Tsimperidis,

Grunova,

Papakostas

2023

IEEE Access

View full text Add to dashboard Cite

Knowing some characteristics of an unknown user is useful information for security and commercial purposes. One of the acquired characteristics is the user's native language, and its recognition can be achieved with data derived from the text he/she types, since text is the most widespread means of communication between Internet users. Keystroke dynamics, which leverages data derived from how user types, ensures that no sensitive data are leaked. In this work, data from the daily typing of users of five different native languages are collected, keystroke dynamics features are extracted, the most suitable ones are selected using a feature selection algorithm, well-known machine learning models and a boosting algorithm are used, and a rate of correct prediction that exceeds 90% is achieved. Knowing a user's native language can help strengthen authentication systems, make better use of online services, and protect unsuspecting users from falling victim to fraud.

show abstract