2,140 Hits in 3.3 sec

Towards Adversarial Phishing Detection

Thomas Kobber Panum, Kaspar Hageman, René Rydhof Hansen, Jens Myrup Pedersen
2020 USENIX Security Symposium  
Over the recent decades, numerous evaluations of automated methods for detecting phishing attacks have been reporting stellar detection performances based on empirical evidence.  ...  These performances often neglect the adaptive behavior of an adversary seeking to evade detection, yielding uncertainty about their adversarial robustness.  ...  The implementation of the experiments for the conducted assessment, including reproductions of detection solutions and perturbation methods, is provided with open access at towards-adversarial-phishing-detection  ... 
dblp:conf/uss/PanumHHP20 fatcat:dkdcaneekrewrgdkzodkhb2w44

An Evasion Attack against ML-based Phishing URL Detectors [article]

Bushra Sabir
2020 arXiv   pre-print
Over the year, Machine Learning Phishing URL classification (MLPU) systems have gained tremendous popularity to detect phishing URLs proactively.  ...  To achieve this, we first develop an algorithm to generate adversarial phishing URLs. We then reproduce 41 MLPU systems and record their baseline performance.  ...  To address these limitations, Machine Learning based Phishing URL detection (MLPU) systems are widely adapted to detect phishing URLs [50, 76, 96] .  ... 
arXiv:2005.08454v1 fatcat:vbl5jdynpjg5thqy2xefi5bizi

DeltaPhish: Detecting Phishing Webpages in Compromised Websites [article]

Igino Corona and Battista Biggio and Matteo Contini and Luca Piras and Roberto Corda and Mauro Mereu and Guido Mureddu and Davide Ariu and Fabio Roli
2017 arXiv   pre-print
DeltaPhish is also robust against adversarial attempts in which the HTML code of the phishing page is carefully manipulated to evade detection.  ...  To the best of our knowledge, this is the first work that specifically leverages this adversarial behavior for detection purposes.  ...  On the other hand, website compromise is only a pivoting step towards the final goal of the phishing scam.  ... 
arXiv:1707.00317v1 fatcat:yrafev7fuzad7ffck6nblsy7z4

Detecting Malicious Accounts showing Adversarial Behavior in Permissionless Blockchains [article]

Rachit Agarwal, Tanmay Thapliyal, Sandeep K. Shukla
2021 arXiv   pre-print
Further, the previously used ML algorithms for identifying malicious accounts show bias towards a particular malicious activity which is over-represented.  ...  In the sequel, we identify that Neural Networks (NN) holds up the best in the face of such bias inducing dataset at the same time being robust against certain adversarial attacks.  ...  This proves the existence of bias in ETC towards 'Phishing'.  ... 
arXiv:2101.11915v1 fatcat:djkwi3p5xnh23bsoqruvtt4stu

On Designing and Evaluating Phishing Webpage Detection Techniques for the Real World

Samuel Marchal, N. Asokan
2018 USENIX Security Symposium  
These guidelines can improve the effectiveness of phishing detection techniques in real-world scenarios and foster technology transfer.  ...  We hope to raise awareness about practices causing this gap and present a set of guidelines for the design and evaluation of phishing webpage detection techniques.  ...  Resilience to adversaries Knowing a detection technique, phishers will try to evade it by adapting their phishes.  ... 
dblp:conf/uss/MarchalA18 fatcat:edhht2iq35cadkww3rpq7c5dwy

URLTran: Improving Phishing URL Detection Using Transformers [article]

Pranav Maneriker, Jack W. Stokes, Edir Garcia Lazo, Diana Carutasu, Farid Tajaddodianfar, Arun Gururajan
2021 arXiv   pre-print
Early phishing detection used standard machine learning classifiers, but recent research has instead proposed the use of deep learning models for the phishing URL detection task.  ...  Browsers often include security features to detect phishing web pages. In the past, some browsers evaluated an unknown URL for inclusion in a list of known phishing pages.  ...  Building URLTran employs a two-pronged approach towards adapting transformers for the task of phishing URL detection.  ... 
arXiv:2106.05256v3 fatcat:uysix2k4kzb6xlyqz7xcahuoym

Beyond the lock icon: real-time detection of phishing websites using public key certificates

Zheng Dong, Apu Kapadia, Jim Blythe, L. Jean Camp
2015 2015 APWG Symposium on Electronic Crime Research (eCrime)  
We further show that this approach works not only against HTTPS-enabled phishing attacks, but also detects HTTP phishing attacks with port 443 enabled.  ...  Other local client-side phishing detection approaches also exist, but primarily rely on page content or URLs, which are arguably easier to manipulate by attackers.  ...  Currently use of valid certificate by adversaries is infrequent; however, attackers are part of the long term trend towards more ubiquitous https.  ... 
doi:10.1109/ecrime.2015.7120795 dblp:conf/ecrime/DongKBC15 fatcat:u3iplbeft5hrdkdwfgoj2et2tu

Characterizing Phishing Threats with Natural Language Processing [article]

Michael C. Kotson, Alexia Schulz
2015 arXiv   pre-print
Spear phishing is a widespread concern in the modern network security landscape, but there are few metrics that measure the extent to which reconnaissance is performed on phishing targets.  ...  In this work we use Natural Language Processing techniques to investigate a specific real-world phishing campaign and quantify attributes that indicate a targeted spear phishing attack.  ...  detected per day fell sharply in the same time range [2] .  ... 
arXiv:1508.07885v1 fatcat:xraqqfl77fhq3elfkel7s2hjuu

VisualPhishNet: Zero-Day Phishing Website Detection by Visual Similarity [article]

Sahar Abdelnabi and Katharina Krombholz and Mario Fritz
2020 arXiv   pre-print
We furthermore present VisualPhish, the largest dataset to date that facilitates visual phishing detection in an ecologically valid manner.  ...  This paper contributes VisualPhishNet, a new similarity-based phishing detection framework, based on a triplet Convolutional Neural Network (CNN).  ...  There is a lot of work towards fixing the evasion problem [3] , however, adversarial perturbations are well-known for classification models.  ... 
arXiv:1909.00300v4 fatcat:ytxzxjswjbcx3eoszjmzcs2jhi

Mobile Application Impersonation Detection Using Dynamic User Interface Extraction [chapter]

Luka Malisa, Kari Kostiainen, Michael Och, Srdjan Capkun
2016 Lecture Notes in Computer Science  
As the detection is based on the visual appearance of the application, as seen by the user, our approach is robust towards the attack implementation technique and resilient to simple detection avoidance  ...  In this paper we present a novel approach for detection of mobile app impersonation attacks.  ...  Phishing detection.  ... 
doi:10.1007/978-3-319-45744-4_11 fatcat:w2m4k4p5xvb53ka3ve5pgliewy

Shadows Behind the Keyboard

Daniel N. Jones
2022 Proceedings of the 2022 ACM on International Workshop on Security and Privacy Analytics  
Through the study of individual differences, we can better understand not only who is most likely to engage in criminal activity in cyberspace but the dispositional tendencies towards specific types of  ...  Specifically, because individuals high in Machiavellianism are more cautious and they make more extensive changes between phishing emails in order to avoid detection and evade spam filters.  ...  Although recapturing a resource is costly, it is far costlier to let a resource remain in the control of an adversary.  ... 
doi:10.1145/3510548.3519379 fatcat:zq6uix6ejfapxkvwyx6adqvhra

Phishing URL Detection Through Top-level Domain Analysis: A Descriptive Approach

Orestis Christou, Nikolaos Pitropakis, Pavlos Papadopoulos, Sean McKeown, William Buchanan
2020 Proceedings of the 6th International Conference on Information Systems Security and Privacy  
Phishing is considered to be one of the most prevalent cyber-attacks because of its immense flexibility and alarmingly high success rate.  ...  This study aims to develop a machine-learning model to detect fraudulent URLs which can be used within the Splunk platform.  ...  For the detection of phishing domains, more focus should have been given towards features explicitly targeting the detection of squatting domains (Moubayed et al., 2018) .  ... 
doi:10.5220/0008902202890298 dblp:conf/icissp/ChristouPPMB20 fatcat:vhjczp26qnerdbr7hi7p7qyq7a

PhiGARo: Automatic Phishing Detection and Incident Response Framework

Martin Husak, Jakub Cegan
2014 2014 Ninth International Conference on Availability, Reliability and Security  
We present a comprehensive framework for automatic phishing incident processing and work in progress concerning automatic phishing detection and reporting.  ...  The honeypots are used to capture e-mails, automatically detect messages containing phishing and immediately transfer them to PhiGARo.  ...  [6] mimicked user responses to detect phishing. Their work places the response before detection to provide the adversary with fake responses.  ... 
doi:10.1109/ares.2014.46 dblp:conf/IEEEares/HusakC14 fatcat:2q637rwncbegtjrbo3irocheva

The Threat of Offensive AI to Organizations [article]

Yisroel Mirsky, Ambra Demontis, Jaidip Kotak, Ram Shankar, Deng Gelei, Liu Yang, Xiangyu Zhang, Wenke Lee, Yuval Elovici, Battista Biggio
2021 arXiv   pre-print
In particular, cyber adversaries can use AI (such as machine learning) to enhance their attacks and expand their campaigns.  ...  For example, how does an AI-capable adversary impact the cyber kill chain? Does AI benefit the attacker more than the defender?  ...  Many email services use machine learning to detect malicious emails. However, adversaries can use adversarial machine learning to evade detection [57, 74, 136, 137] .  ... 
arXiv:2106.15764v1 fatcat:zkfukg4krjcczpie2gbdznwqqi

Tree-classification Algorithm to Ease User Detection of Predatory Hijacked Journals: Empirical Analysis of Journal Metrics Rankings

Arnold Adimabua Ojugo, Department of Computer Science, Federal University of Petroleum Resources Effurun, Delta State, Nigeria., Obinna Nwankwo
2021 International Journal of Engineering and Manufacturing  
Results show the classification algorithm can effectively detect 95-percent accuracy of journal phishing based on journal metric indicators and website ranks.  ...  The negative impact thus, of predatory and hijacked journals cannot be over-emphasized as adversaries use carefully crafted, social engineering (phishing attack) skillsto exploit unsuspecting and inexperienced  ...  .  Sequence in search result adds to increased accuracy in detecting phishing pages.  ... 
doi:10.5815/ijem.2021.04.01 fatcat:63gtt5vqnveq3muwys7f6d3ho4
« Previous Showing results 1 — 15 out of 2,140 results