Proceedings 2018 Network and Distributed System Security Symposium 2018
DOI: 10.14722/ndss.2018.23303
|View full text |Cite
|
Sign up to set email alerts
|

K-means++ vs. Behavioral Biometrics: One Loop to Rule Them All

Abstract: Abstract-Behavioral biometrics, a field that studies patterns in an individual's unique behavior, has been researched actively as a means of authentication for decades. Recently, it has even been adopted in many real world scenarios. In this paper, we study keystroke dynamics, the most researched of such behavioral biometrics, from the perspective of an adversary. We designed two adversarial agents with a standard accuracy convenience tradeoff: Targeted K-means++, which is an expensive, but extremely effective… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

0
7
0

Year Published

2019
2019
2023
2023

Publication Types

Select...
5
2
2

Relationship

0
9

Authors

Journals

citations
Cited by 19 publications
(7 citation statements)
references
References 20 publications
(32 reference statements)
0
7
0
Order By: Relevance
“…As suggested by Stylios et al (2021b), CA systems need to be evaluated under the high effort approaches to see the actual performance of machine learning and deep learning models under the spectrum of today's possible threats. Therefore, our system should be evaluated against the frog-boiling attack (Wang et al , 2012), the algorithmic attack (Serwadda and Phoha, 2013), the mimic attacks (Negi et al , 2018; Meng et al , 2013) and the snoop-forge-replay attack (Rahman et al , 2013). Finally, our system was tested in a sample of 39 individuals, and we plan to evaluate it in a larger sample of users.…”
Section: Discussionmentioning
confidence: 99%
“…As suggested by Stylios et al (2021b), CA systems need to be evaluated under the high effort approaches to see the actual performance of machine learning and deep learning models under the spectrum of today's possible threats. Therefore, our system should be evaluated against the frog-boiling attack (Wang et al , 2012), the algorithmic attack (Serwadda and Phoha, 2013), the mimic attacks (Negi et al , 2018; Meng et al , 2013) and the snoop-forge-replay attack (Rahman et al , 2013). Finally, our system was tested in a sample of 39 individuals, and we plan to evaluate it in a larger sample of users.…”
Section: Discussionmentioning
confidence: 99%
“…To manage expansive datasets, significant exertion has likewise gone into additionally accelerating k-means, most strikingly by utilizing KD-trees or abusing the triangular disparity to abstain from contrasting every data point and every one of the centroids amid the task step. Persistent upgrades and speculations of the fundamental algorithm have guaranteed it's proceeded with pertinence and step by step expanded its adequacy too [14].…”
Section: Generalizations and Communicationsmentioning
confidence: 99%
“…For example, a masterface [37], [46] is a face image that matches a surprisingly large part of the population, making it an exciting candidate to break a face recognition system. Moreover, this concept is not constrained to the face domain; examples exist for fingerprints [14] and keystrokes [36], [44]. The fact that biometric recognition performance is not uniform among individuals was already demonstrated in 1998 by Doddington [20].…”
Section: Introductionmentioning
confidence: 99%