Zero-day Malicious Email Investigation and Detection Using Features with Deep-learning Approach

Phomkeona, Sanouphab; Okamura, Koji

doi:10.2197/ipsjjip.28.222

Cited by 5 publications

(4 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The additional enhancement will soon be needed since the development in hardware is slower than the improvement in DL computing power, which limits DL models' performance. Furthermore, complex DL models using GPUs and TPUs in their implementation have specific 2 Alexa https://www.alexa.com/ Amazon - [21], [33], [43], [45], [48], [77], [96], [97], [99], [101], [105] 3 DMOZ https://dmoz-odp.org/ https://dmoztools.net/docs/en/rdf.html Open Directory Project - [43], [58], [82], [88], [91], [93], [ effects on the environment and energy consumption. Carbon dioxide emitted from such models is approximately five times an average car's lifetime emission.…”

Section: ) Computational Constraintsmentioning

confidence: 99%

See 1 more Smart Citation

Deep Learning for Phishing Detection: Taxonomy, Current Challenges and Future Directions

Selamat

Krejcar

et al. 2022

IEEE Access

View full text Add to dashboard Cite

Phishing has become an increasing concern and captured the attention of end-users as well as security experts. Despite decades of development and improvement, existing phishing detection techniques still suffer from the deficiency in performance accuracy and the inability to detect unknown attacks. Motivated to solve these problems, many researchers in the cybersecurity domain have shifted their attention to phishing detection that capitalizes on machine learning techniques. In recent years, deep learning has emerged as a branch of machine learning that has become a promising solution for phishing detection. As a result, this study proposes a taxonomy of deep learning algorithms for phishing detection by examining 81 selected papers using a systematic literature review approach. The paper first introduces the concept of phishing and deep learning in the context of cybersecurity. Then, phishing detection and deep learning algorithm taxonomies are provided to classify the existing literature into various categories. Next, taking the proposed taxonomy as a baseline, this study comprehensively reviews the state-of-the-art deep learning techniques and analyzes their advantages as well as disadvantages. Subsequently, the paper discusses various issues deep learning faces in phishing detection and proposes future research directions to overcome these challenges. Finally, an empirical analysis is conducted to evaluate the performance of various deep learning techniques in a practical context and highlight the related issues that motivate researchers in their future works. The results obtained from the empirical experiment showed that the common issues among most of the state-of-the-art deep learning algorithms are manual parameter-tuning, long training time, and deficient detection accuracy.

show abstract

Section: ) Computational Constraintsmentioning

confidence: 99%

“…DL requires a significant amount of data and time to train the model [103]. Datasets used for training neural networks usually contain millions of samples [44], [45], [81], [88], [98]. As a result, they need a longer time to train the model to obtain high-performance accuracy.…”

Section: ) Time Complexitymentioning

confidence: 99%

Deep Learning for Phishing Detection: Taxonomy, Current Challenges and Future Directions

Selamat

Krejcar

et al. 2022

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Phomkeona and Okamura [ 16 ] proposed a method for categorizing and diagnosing zero-day malicious emails based on data gathered from the email header and content. They merged it with dynamic analytic data as a collection of 27 features, including machine translation detection, risk word detection, and other characteristics, using numerous application programming interfaces.…”

Section: Literature Reviewmentioning

confidence: 99%

An Advanced Deep Attention Collaborative Mechanism for Secure Educational Email Services

Chen¹,

Yang²

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

The COVID-19 crisis has once again highlighted the vulnerabilities of some critical areas in cyberspace, especially in the field of education, as distance learning and social distance have increased their dependence on digital technologies and connectivity. Many recent cyberattacks on e-learning systems, educational content services, and trainee management systems have created severe demands for specialized technological solutions to protect the security of modern training methods. Email is one of the most critical technologies of educational organizations that are attacked daily by spam, phishing campaigns, and all kinds of malicious programs. Considering the efforts made by the global research community to ensure educational processes, this study presents an advanced deep attention collaborative filter for secure academic email services. It is a specialized application of intelligent techniques that, for the first time, examines and models the problem of spam as a system of graphs where collaborative referral systems undertake the processing and analysis of direct and indirect social information to detect and categorize spam emails. In this study, nonnegative matrix factorization (NMF) is applied to the social graph adjacent table to place users in one (or more) overlapping communities. Also, using a deep attention mechanism, it becomes personalized for each user. At the same time, with the introduction of exponential random graph models (ERGMs) in the process of factorization, local dependencies are significantly mitigated to achieve the revelation of malicious communities. This methodology is being tested successfully in implementing mail protection systems for educational organizations. According to the findings, the proposed algorithm outperforms all other compared algorithms in every metric tested.

show abstract

“…Sanouphab Phomkeona and Okamura [12] proposed a new method to extract features from email and a deep-learning approach to detect zero-day malware spam. They extracted some features from e-mail's header and body parts that included risk words detected, machine translation detected, and other features by using several APIs.…”

Section: Rastenis Et Al (2021)mentioning

confidence: 99%

Evaluation of Machines Learning Algorithms in Detection of Malware-based Phishing Attacks for Securing E-Mail Communication

Kisambu¹,

Mjahidi²

2022

Artificial Intelligence and Machine Learning

View full text Add to dashboard Cite

Malicious software, commonly known as Malware is one of the most significant threats facing Internet users today. Malware-based phishing attacks are among the major threats to Internet users that are difficult to defend against because they do not appear to be malicious in nature. There were several initiatives in combating phishing attacks but there are many difficulties and obstacles encountered. This study deals with evaluation of machine learning algorithms in detection of malware-based phishing attacks for securing email communication. It deeply evaluate the efficacy of the algorithms when integrated with major open-source security mail filters with different mitigation techniques. The main classifiers used such as SVM, KNN, Logistic Regression and Naïve Bayes were evaluated using performance metrics namely accuracy, precision, recall and f-score. Based on the findings, the study proposed improvement for securing e-mail communication against malware-based phishing using the best performing machine-learning algorithm to keep pace with malware evolution.

show abstract

Zero-day Malicious Email Investigation and Detection Using Features with Deep-learning Approach

Cited by 5 publications

References 13 publications

Deep Learning for Phishing Detection: Taxonomy, Current Challenges and Future Directions

Deep Learning for Phishing Detection: Taxonomy, Current Challenges and Future Directions

An Advanced Deep Attention Collaborative Mechanism for Secure Educational Email Services

Evaluation of Machines Learning Algorithms in Detection of Malware-based Phishing Attacks for Securing E-Mail Communication

Contact Info

Product

Resources

About