Learning to detect phishing emails

Fette, Ian; Sadeh, Norman; Tomasic, Anthony

doi:10.1145/1242572.1242660

Cited by 400 publications

(204 citation statements)

References 7 publications

Supporting

Mentioning

203

Contrasting

Order By: Relevance

“…Most research about protecting users from phishing emails is about methods to check manually or automatically for keywords, grammatical inconsistencies, typos, or information misplacement [8][9][10]. All these methods can help the user, but they have not yet changed the fact that people are still falling for phishing emails [10].…”

Section: Motivationmentioning

confidence: 99%

Principles of Persuasion in Social Engineering and Their Use in Phishing

Ferreira

Coventry

Lenzini

2015

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Research on marketing and deception has identified principles of persuasion that influence human decisions. However, this research is scattered: it focuses on specific contexts and produces different taxonomies. In regard to frauds and scams, three taxonomies are often referred in the literature: Cialdini's principles of influence, Gragg's psychological triggers, and Stajano et al. principles of scams. It is unclear whether these relate but clearly some of their principles seem overlapping whereas others look complementary. We propose a way to connect those principles and present a merged and reviewed list for them. Then, we analyse various phishing emails and show that our principles are used therein in specific combinations. Our analysis of phishing is based on peer review and further research is needed to make it automatic, but the approach we follow, together with principles we propose, can be applied more consistently and more comprehensively than the original taxonomies.

show abstract

Section: Motivationmentioning

confidence: 99%

Principles of Persuasion in Social Engineering and Their Use in Phishing

Ferreira

Coventry

Lenzini

2015

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…It is often easier for attackers to exploit human and social weaknesses of the defences than to defeat the technological countermeasures [18]. This is also evident in anti-phishing literature as most research focused on technical solutions such as: developing browser toolbars/plug-ins [23] preventative measures, characteristics and email structure [6], [20], [22], algorithms for detecting, identifying and measuring phishing emails and sites [8], [11], [32] and evaluating the effectiveness of web browser toolbar warnings/indicators [4], [7], [12], [31]. Many employees cannot identify the difference between a genuine and a spoofed website [4], [21].…”

Section: The Need For a Holistic Anti-phishing Frameworkmentioning

confidence: 99%

An Enterprise Anti-phishing Framework

Frauenstein¹,

Solms²

2013

Information Assurance and Security Education and Training

View full text Add to dashboard Cite

Abstract. The objective of this paper is to report back on an organizational framework, which consisted of human, organization and technology (HOT) dimensions in holistically addressing aspects associated with phishing. Most anti-phishing literature studied either focused on technical controls or education in isolation however; education is core to all aspects in the above-mentioned framework. It is evident, from literature, that little work has been conducted on anti-phishing preventative measures in the context of organizations but rather from a personal user-level. In the framework, the emphasis is placed on the human factors in addressing phishing attacks.

show abstract

“…PFILTER, which was proposed by Fette et al [8], employed SVM to distinguish phishing emails from other emails. According to [9], Abu-Nimeh et al compared the predictive accuracy of several machine learning methods including LR, CART, RF, NB, SVM, and BART.…”

Section: Related Workmentioning

confidence: 99%

An Evaluation of Machine Learning-Based Methods for Detection of Phishing Sites

Miyamoto

Hazeyama

Kadobayashi

2009

Advances in Neuro-Information Processing

View full text Add to dashboard Cite

Abstract. In this paper, we evaluate the performance of machine learningbased methods for detection of phishing sites. In our previous work [1], we attempted to employ a machine learning technique to improve the detection accuracy. Our preliminary evaluation showed the AdaBoost-based detection method can achieve higher detection accuracy than the traditional detection method. Here, we evaluate the performance of 9 machine learning techniques including AdaBoost, Bagging, Support Vector Machines, Classification and Regression Trees, Logistic Regression, Random Forests, Neural Networks, Naive Bayes, and Bayesian Additive Regression Trees. We let these machine learning techniques combine heuristics, and also let machine learning-based detection methods distinguish phishing sites from others. We analyze our dataset, which is composed of 1,500 phishing sites and 1,500 legitimate sites, classify them using the machine learning-based detection methods, and measure the performance. In our evaluation, we used f 1 measure, error rate, and Area Under the ROC Curve (AUC) as performance metrics along with our requirements for detection methods. The highest f1 measure is 0.8581, the lowest error rate is 14.15%, and the highest AUC is 0.9342, all of which are observed in the case of AdaBoost. We also observe that 7 out of 9 machine learning-based detection methods outperform the traditional detection method.

show abstract

Learning to detect phishing emails

Cited by 400 publications

References 7 publications

Principles of Persuasion in Social Engineering and Their Use in Phishing

Principles of Persuasion in Social Engineering and Their Use in Phishing

An Enterprise Anti-phishing Framework

An Evaluation of Machine Learning-Based Methods for Detection of Phishing Sites

Contact Info

Product

Resources

About