A Survey on Keystroke Dynamics Biometrics: Approaches, Advances, and Evaluations

Zhong, Yu; Deng, Yunbin

doi:10.15579/gcsr.vol2.ch1

Cited by 34 publications

(21 citation statements)

References 100 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Keystroke dynamic features can be extracted using the timing information of the key press and release events. The hold time of individual keys and the latency between keys (the time interval between pressing one key and a succeeding key) are typically exploited [ 20 ]. In addition to ordered pairs (two successive keystrokes), n-tuples of a sequence of keystrokes have also been investigated and keystroke biometrics research has utilised many machine learning and classification techniques.…”

Section: Introductionmentioning

confidence: 99%

High-accuracy detection of early Parkinson's Disease using multiple characteristics of finger movement while typing

Adams

2017

PLoS ONE

View full text Add to dashboard Cite

Parkinson’s Disease (PD) is a progressive neurodegenerative movement disease affecting over 6 million people worldwide. Loss of dopamine-producing neurons results in a range of both motor and non-motor symptoms, however there is currently no definitive test for PD by non-specialist clinicians, especially in the early disease stages where the symptoms may be subtle and poorly characterised. This results in a high misdiagnosis rate (up to 25% by non-specialists) and people can have the disease for many years before diagnosis. There is a need for a more accurate, objective means of early detection, ideally one which can be used by individuals in their home setting. In this investigation, keystroke timing information from 103 subjects (comprising 32 with mild PD severity and the remainder non-PD controls) was captured as they typed on a computer keyboard over an extended period and showed that PD affects various characteristics of hand and finger movement and that these can be detected. A novel methodology was used to classify the subjects’ disease status, by utilising a combination of many keystroke features which were analysed by an ensemble of machine learning classification models. When applied to two separate participant groups, this approach was able to successfully discriminate between early-PD subjects and controls with 96% sensitivity, 97% specificity and an AUC of 0.98. The technique does not require any specialised equipment or medical supervision, and does not rely on the experience and skill of the practitioner. Regarding more general application, it currently does not incorporate a second cardinal disease symptom, so may not differentiate PD from similar movement-related disorders.

show abstract

Section: Introductionmentioning

confidence: 99%

High-accuracy detection of early Parkinson's Disease using multiple characteristics of finger movement while typing

Adams

2017

PLoS ONE

View full text Add to dashboard Cite

show abstract

“…Digraph latencies within the virtual letter pairs were then used to produce a score [3]. Features like digraphs and trigraphs rely only on the word context [4]. To solve the [5] used digraphs, trigraphs and n-graphs features along with the keyword latencies i.e.…”

Section: Literature Surveymentioning

confidence: 99%

Non-Conventional Factors for Keystroke Dynamics as a Support Factor for Authenticating User

Raul*¹,

Shankarmani²

2020

IJITEE

View full text Add to dashboard Cite

Keystroke Dynamics Are Most Frequently UsedIn Situation Where Password Security Is The Major Concern. In The Event That The User’s Secret Word Is Disclosed, AndThe Keystroke Rhythm Of The Genuine Client Is Known, The Application Will Not Accept The Fake User In Spite Of Having Valid Credentials. This Paper Focuses On Strengthening The Keystroke Dynamics By Introducing Non-Conventional Features For Authenticating A User Using Static Keystroke Dynamics. The Paper Also Discusses The Improvement In Far, Frr, Err, And Accuracy After The Application Of The Proposed NonConventional Feature Set Along With The Existing Timing Feature Set.

show abstract

“…This difference in std distance suggest that good users tend to have less keystrokes variations. Table 1 (Left) were calculated independently for each of the 300 subjects (300 different decision thresholds), these EERs are calculated as the average of the individual EER from all subjects [3,10,12]. To analyse the impact of the score normalization in the performance, average EER from the whole database (using only one decision threshold for all users) are summarized in Table 3.…”

Section: Stability Of the Featuresmentioning

confidence: 99%

“…Given the wide range of potential practical applications mentioned above, a heterogeneous community of researchers from different fields has produced in the last decade a very large number of works studying different aspects of keystroke recognition. Those contributions have been compiled in several surveys [2,3,4,5] that describe the technology in terms of performance, databases, privacy and security. The techniques are usually divided into fixed text (the text used to model the typing behavior of the user and the text used to authenticate is the same) and free text (the text used to model the typing behavior and the text used to authenticate do not necessarily match).…”

Section: Introductionmentioning

confidence: 99%

On the Analysis of Keystroke Recognition Performance based on Proprietary Passwords

Acien

Hernandez-Ortega

Morales

et al. 2017

8th International Conference of Pattern Recognition Systems (ICPRS 2017)

View full text Add to dashboard Cite

This paper investigates factors affecting the keystroke recognition performance. Keystroke recognition has attracted the interest of scientific community because of the many challenges associated to modelling the typing behavior. Despite the great efforts made during the last decades, the performance of keystroke recognition systems is far from the performance achieved by traditional hard biometrics. This is very pronounced for some users, who generate many recognition errors even with the most sophisticate recognition algorithms. Our purpose here is to study factors affecting the performance of users for scenarios in which each user employ a proprietary password based on familiar information. The experiments comprise a public database with 300 users (300 passwords) and four state-of-the-art recognition systems recently evaluated during the Keystroke Biometrics Ongoing Competition. The results suggest the importance of the correct alignment of samples and intra-class variability despite the impact of the length of the password and timing features.

show abstract

A Survey on Keystroke Dynamics Biometrics: Approaches, Advances, and Evaluations

Cited by 34 publications

References 100 publications

High-accuracy detection of early Parkinson's Disease using multiple characteristics of finger movement while typing

High-accuracy detection of early Parkinson's Disease using multiple characteristics of finger movement while typing

Non-Conventional Factors for Keystroke Dynamics as a Support Factor for Authenticating User

On the Analysis of Keystroke Recognition Performance based on Proprietary Passwords

Contact Info

Product

Resources

About