Email phishing: text classification using natural language processing

Verma, Priyanka; Goyal, Anjali; Gigras, Yogita

doi:10.11591/csit.v1i1.p1-12

Cited by 25 publications

(20 citation statements)

References 7 publications

(7 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One of the main focuses of recent research has been Natural Language Processing (NLP), this technique will allow for better detection and filtering of phishing emails that implement semantic changes to get past existing filters. Research suggests that this technique is more accurate in filtering out phishing emails than existing techniques, although these methods are yet to be implemented on larger datasets [90,91]. Neural networks are also being considered as methods of phishing prevention; however, these are often criticized as they require long training periods and the knowledge of experts to tune the parameters.…”

Section: Discussion Of Current Challenges and Trends In Phishing Attacksmentioning

confidence: 99%

Phishing Attacks Survey: Types, Vectors, and Technical Approaches

Alabdan

2020

Future Internet

117

View full text Add to dashboard Cite

Phishing attacks, which have existed for several decades and continue to be a major problem today, constitute a severe threat in the cyber world. Attackers are adopting multiple new and creative methods through which to conduct phishing attacks, which are growing rapidly. Therefore, there is a need to conduct a comprehensive review of past and current phishing approaches. In this paper, a review of the approaches used during phishing attacks is presented. This paper comprises a literature review, followed by a comprehensive examination of the characteristics of the existing classic, modern, and cutting-edge phishing attack techniques. The aims of this paper are to build awareness of phishing techniques, educate individuals about these attacks, and encourage the use of phishing prevention techniques, in addition to encouraging discourse among the professional community about this topic.

show abstract

Section: Discussion Of Current Challenges and Trends In Phishing Attacksmentioning

confidence: 99%

Phishing Attacks Survey: Types, Vectors, and Technical Approaches

Alabdan

2020

Future Internet

117

View full text Add to dashboard Cite

show abstract

“…In [1], the phishing emails are classified based on machine learning approach. For this, "The Short Message Service Spam Collection v.1" dataset is used.…”

Section: Related Workmentioning

confidence: 99%

“…Classifiers such as Decision Tree, SVC, Random Forest and K-Nearest Neighbor are used for spam and ham classification. This method is implemented in python using Anaconda Jupyter lab [1]. The spam mails are traditionally classified using supervised classification model.…”

Section: Related Workmentioning

confidence: 99%

“…This network threat is most commonly done by sending malicious links via e-mails or by sending banners and pop-ups which can be clicked without the conscious of the user and hence they could get trapped. Phishing is classified as Spear phishing, in which an individual or a group of individuals are targeted for phishing attack, Clone phishing, in which a clone of an existing email with a malicious link is sent to any of the users, since the target is only stealing the credential of everyone, and not for a specific person, Whaling, in which the higher level or senior executives like CEO of an organization are targeted for phishing attack [1].Abuse of Online Social Networks (OSN) is found frequently in news, like Facebook Cambridge Analytica crisis and bot-based election manipulation in Twitter. The media are only involved recently in abuse of OSN, but it has been happening a long decade.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Detection of Email Spam using Natural Language Processing Based Random Forest Approach

Nivedha¹,

Raja²

2022

IJCSMC

View full text Add to dashboard Cite

An unsolicited means of digital communications in the internet world is the spam email, which could be sent to an individual or a group of individuals or a company. These spam emails may cause serious threat to the user i.e., the email addresses used for any online registrations may be collected by the malignant third parties (spammers) and they expose the genuine user to various kinds of attacks. Another method of spamming is by creating a temporary email register and receive emails that can be terminated after some certain amount of time. This method is well suited for misusing those temporary email addresses for sending free spam emails without revealing the spammers real account details. These attacks create major problems like theft of user credentials, lack of storage, etc. Hence it is essential to introduce an efficient detection mechanism through feature extraction and classification for detecting spam emails and temporary email addresses. This can be accomplished through a novel Natural Language Processing based Random Forest (NLP-RF) approach. With the help of our proposed approach, the spam emails are reduced and this method improves the accuracy of spam email filtering, since the use of NLP makes the system to detect the natural languages spoken by people and the Random Forest approach uses multiple decision trees and uses a random node for filtering the spams.

show abstract

“…They reached a very high classification accuracy of 99.84%. Verma et al (2020) looked at the classification of phishing emails using NLP.…”

Section: Related Workmentioning

confidence: 99%

Machine Learning and Deep Learning for Phishing Email Classification using One-Hot Encoding

Nandi¹,

Bagui²,

White³

2021

Journal of Computer Science

View full text Add to dashboard Cite

Representation of text is a significant task in Natural Language Processing (NLP) and in recent years Deep Learning (DL) and Machine Learning (ML) have been widely used in various NLP tasks like topic classification, sentiment analysis and language translation. Until very recently, little work has been devoted to semantic analysis in phishing detection or phishing email detection. The novelty of this study is in using deep semantic analysis to capture inherent characteristics of the text body. One-hot encoding was used with DL and ML techniques to classify emails as phishing or nonphishing. A comparison of various parameters and hyperparameters was performed for DL. The results of various ML models, Naïve Bayes, SVM, Decision Tree, as well as DL models, Convolutional Neural Networks (CNN) and Long Short Term Memory (LSTM), were presented. The DL models performed better than the ML models in terms of accuracy, but the ML models performed better than the DL models in terms of computation time. CNN with Word Embedding performed the best in terms of accuracy (96.34%), demonstrating the effectiveness of semantic analysis in phishing email detection.

show abstract

Email phishing: text classification using natural language processing

Cited by 25 publications

References 7 publications

Phishing Attacks Survey: Types, Vectors, and Technical Approaches

Phishing Attacks Survey: Types, Vectors, and Technical Approaches

Detection of Email Spam using Natural Language Processing Based Random Forest Approach

Machine Learning and Deep Learning for Phishing Email Classification using One-Hot Encoding

Contact Info

Product

Resources

About