Email Classification Research Articles

Deep Neural Network Based Spam Email Classification Using Attention Mechanisms

Deep Neural Network Based Spam Email Classification Using Attention Mechanisms

Spam email classification and sentiment analysis based on semantic similarity methods

Spam email classification and sentiment analysis based on semantic similarity methods

FEATURE ENGINEERING WITH SENTENCE SIMILARITY USING THE LONGEST COMMON SUBSEQUENCE FOR EMAIL CLASSIFICATION

Feature selection plays a prominent role in email classification since selecting the most relevant features enhances the accuracy and performance of the learning classifier. Due to the exponential increase rate in the usage of emails, the classification of such emails posed a fitting problem. Therefore, there is a requirement for a proper classification system. Such an email classification system requires an efficient feature selection method for the accurate classification of the most relevant features. This paper proposes a novel feature selection method for sentence similarity using the longest common subsequence for email classification. The proposed feature selection method works in two main phases: First, it builds the longest common subsequence vector of features by comparing each email with all other emails in the dataset. Later, a template is constructed for each class using the closest features of emails of a particular class. Further, email classification is tested for unseen emails using these templates. The performance of the proposed method is compared with traditional feature selection methods such as TF-IDF, Information Gain, Chi-square, and semantic approach. The experimental results showed that the proposed method performed well with 96.61% accuracy.

THE EFFECTIVENESS OF URL FEATURES ON PHISHING EMAILS CLASSIFICATION USING MACHINE LEARNING APPROACH

Phishing email classification requires features so that the performance obtained produces good accuracy. One of the reasons for the lack of development of models for detecting phishing emails is the complexity of the feature selection. Feature selection is one of the essential parts of getting a good classification result, commonly used features are header, body, and Uniform Resource Locator (URL). Besides the email body text content, the URL is one of the leading indicators that the phishing attack successfully happened. The URL is commonly located on the body of the phishing email to get the victim's attention. It will redirect the victim to a fake website to obtain personal information from the victim. There is a lack of information about how the URL features affect the phishing email classification results. Therefore, this work focuses on using URL features to determine whether an email is phishing or legitimate using machine learning approaches. Two public datasets used in this work are the Online Phishing Corpus and Enron Corpus. The URL features are extracted using the Beautiful Soup library. Two machine learning classifiers used in this work are Support Vector Machine (SVM) and Artificial Neural Network (ANN). The experiments were divided into two based on features used in the classifiers. The first experiment used raw email data with URL features, while the second only used raw email data. The first experiment shows higher accuracy in both classifiers, SVM and ANN. Hence, this research proves that the impact of selecting URL features will increase the performance of the classification.

Privacy-preserving spam filtering using homomorphic and functional encryption

Conventional spam classification requires the end-users to reveal the content of incoming emails to a classifier so that text analysis can be performed. On the other hand, new cryptographic primitives allow this classification task to be performed on encrypted emails without revealing the email contents, hence preserves user data privacy. In this paper, we construct a spam classification framework that enables the classification of encrypted emails. Our model is based on a neural network with a quadratic network component and a multi-layer perceptron network component. The quadratic network architecture is compatible with the operation of an existing quadratic functional encryption scheme. To protect email content privacy, we proposed two spam classification solutions based on homomorphic encryption (HE) and functional encryption (FE) that enables our classifiers to predict the label of encrypted emails. The evaluation results on real-world spam datasets indicate that our proposed spam classification solutions achieve accuracies over 95%. Our performance study and security analysis provide pros and cons of each proposed solution. For instance, the FE solution predicts a label of an encrypted email in less than 31 s whereas the HE solution takes up to 265 s to do so. Nonetheless, the HE solution is not prone to potential information leakage as the FE solution.

A Comparison of Machine Learning Models to Prioritise Emails using Emotion Analysis for Customer Service Excellence

There has been little research on machine learning for email prioritization for customer service excellence. To fill this gap, we propose and assess the efficacy of various machine learning techniques for classifying emails into three degrees of priority: high, low, and neutral, based on the emotions inherent in the email content. It is predicted that after emails are classified into those three categories, recipients will be able to respond to emails more efficiently and provide better customer service. We use the NRC Emotion Lexicon to construct a labeled email dataset of 517,401 messages for our proposal. Following that, we train and test four prominent machine learning models, MNB, SVM, LogR, and RF, and an Ensemble of MNB, LSVC, and RF classifiers, on the labeled dataset. Our main findings suggest that machine learning may be used to classify emails based on their emotional content. However, some models outperform others. During the testing phase, we also discovered that the LogR and LSVC models performed the best, with an accuracy of 72%, while the MNB classifier performed the poorest. Furthermore, classification performance differed depending on whether the dataset was balanced or imbalanced. We conclude that machine learning models that employ emotions for email classification are a promising avenue that should be explored further.

Why people keep falling for phishing scams: The effects of time pressure and deception cues on the detection of phishing emails

Lack of personalisation and poor mechanics (e.g., grammar, spelling and punctuation) are commonly cited as cues of deception that people can use to identify phishing emails. However, in an online email classification experiment (N = 472), we found no empirical evidence that the presence of these features was associated with better phishing email discrimination. We also manipulated time pressure and found that it significantly reduced detection accuracy. Participants rarely inspected the URLs associated with links in the phishing emails but, when they did, their detection performance improved. Better performance in distinguishing between genuine and phishing emails was linked to lower levels of an intuitive decision-making style and relatively lower education levels amongst a highly educated sample. Older participants and those with greater computer proficiency and stronger email habit showed a slight increase in tendency to judge emails as suspicious. The results are discussed in terms of intervention strategies such as cyber security training to improve resilience to phishing attacks.

Performance Evaluation of Machine Learning Algorithms on Textual Datasets for Spam Email Classification

Abstract: Email is one of the most popular modes of communication we have today. Billions of emails are sent every day in our world but not every one of them is relevant or of importance. The irrelevant and unwanted emails are termed email spam. These spam emails are sent with many different targets that range from advertisement to data theft. Filtering these spam emails is very essential in order to keep the email space fluent in its functioning. Machine Learning algorithms are being extensively used in the classification of spam emails. This paper showcases the performance evaluation of some selected supervised Machine Learning algorithms namely Naive Bayes Classifier, Support Vector Machine, Random Forest, & XG-Boost for spam email classification on a combination of three different datasets. For feature extraction, both Bag of Words & TF-IDF models were used separately and performance with both of these approaches was also compared. The results showed that SVM performed better than all the other algorithms when trained with TF-IDF feature vectors. The performance metrics used were accuracy, precision, recall, and f1-score, along with the ROC curve.

Spam Email Classification by Hybrid Feature Selection with Advanced Machine learning Algorithm – Future Perspective

Recently, email has become a common way for people to communicate and share information both officially and personally. Email may be used by spammers to transmit harmful materials to Internet users. The data must be protected from unauthorized access, which necessitates the development of a reliable method for identifying spam emails. As a result, a variety of solutions have been devised. An innovative hybrid machine learning strategy for effectively detecting spam emails has been discussed in this study. This means that identifying spam and non-spam email is a difficult process. Spam email categorization has undergone a significant evolution in recent years, as shown by the research given below. For locating spam, this study uses a mixed approach. Different email categorization algorithms are used to rank them for future perspective.

Classification of Unwanted E-Mails (Spam) with Turkish Text by Different Algorithms in Weka Program

Abstract: Today, with the widespread use of the Internet, electronic communication tools have also been widely used. One of these tools is e-mails. E-mails are easy to use and provide the opportunity to reach thousands of people at the same time. This advantage causes some bad uses. E-mail users are faced with dozens of unsolicited mails (spam) against their will. In this study, 1017 mails collected from about 20 different Gmail and Hotmail accounts were classified as spam or regular e-mail using the algorithms in the Weka program, and the success of the algorithms was compared. In the study, 45 different algorithms were tested. The highest classification success was obtained with the NavieBayesMultinominal and NavieBayesMultinominalUpdateable algorithms with 94.7886% correct classification. Among other classifier algorithms, Trees RandomForest algorithm 93.6087%, Meta. MultiClassClassifier and Functions SGD 92.4287%, Functions SMO 91.7404%, Meta RandomCommittee 91.0521%, Bayes NavieBayes and Bayes NavieBayesUpdateable 90.3638% classification success.

An Integrated Approach of Proposed Pruning Based Feature Selection Technique (PBFST) for Phishing E-mail Detection

Introduction: The entire world is shifting towards electronic communication through Email for fast and secure communication. Millions of people, including organization, government, and others, are using Email services. This growing number of Email users are facing problems; therefore, detecting phishing Email is a challenging task, especially for non-IT users. Automatic detection of phishing Email is essential to deploy along with Email software. Various authors have worked in the field of phishing Email classification with different feature selection and optimization techniques for better performance. Objective: This paper attempts to build a model for the detection of phishing Email using data mining techniques. This paper's significant contribution is to develop and apply Feature Selection Technique (FST) to reduce features from the phishing Email benchmark data set. Methods: The proposed Pruning Based Feature Selection Technique (PBFST) is used to determine the rank of feature based on the level of the tree where feature exists. The proposed algorithm is integrated with already developed Bucket Based Feature Selection Technique (BBFST). BBFST is used as an internal part to rank features in a particular level of the tree. Results : Experimental work was carried out with open source WEKA data mining software using a 10-fold cross-validation technique. The proposed FST was compared with other ranking based FSTs to check the performance of C4.5 classifier with Phishing Email data set. Conclusion: The proposed FST reduces 33 features out of 47 features which exist in phishing Email dataset and C4.5 algorithm produces remarkable accuracy of 99.06% with only 11 features and it has been found to be better than other existing FSTs.

Email classification analysis using machine learning techniques

PurposeIn this digital era, email is the most pervasive form of communication between people. Many users become a victim of spam emails and their data have been exposed.Design/methodology/approachResearchers contribute to solving this problem by a focus on advanced machine learning algorithms and improved models for detecting spam emails but there is still a gap in features. To achieve good results, features also play an important role. To evaluate the performance of applied classifiers, 10-fold cross-validation is used.FindingsThe results approve that the spam emails are correctly classified with the accuracy of 98.00% for the Support Vector Machine and 98.06% for the Artificial Neural Network as compared to other applied machine learning classifiers.Originality/valueIn this paper, Point-Biserial correlation is applied to each feature concerning the class label of the University of California Irvine (UCI) spambase email dataset to select the best features. Extensive experiments are conducted on selected features by training the different classifiers.

Classification of Phishing Email Using Word Embedding and Machine Learning Techniques

Email phishing is a cyber-attack, bringing substantial financial damage to corporate and commercial organizations. A phishing email is a special type of spamming, used to trick the user to disclose personal information to access his digital assets. Phishing attack is generally triggered by emailing links to spoofed websites that collect sensitive information. The APWG survey suggests that the existing countermeasures remain ineffective and insufficient for detecting phishing attacks. Hence there is a need for an efficient mechanism to detect phishing emails to provide better security against such attacks to the common user. The existing open-source data sets are limited in diversity, hence they do not capture the real picture of the attack. Hence there is a need for real-time input data set to design accurate email anti-phishing solutions. In the current work, it has been created a real-time in-house corpus of phishing and legitimate emails and proposed efficient techniques to detect phishing emails using a word embedding and machine learning algorithms. The proposed system uses only four email header-based heuristics for the classification of emails. The proposed word embedding cum machine learning framework comprises six word embedding techniques with five machine learning classifiers to evaluate the best performing combination. Among all six combinations, Random Forest consistently performed the best with FastText (CBOW) by achieving an accuracy of 99.50% with a false positive rate of 0.053%, TF-IDF achieved an accuracy of 99.39% with a false positive rate of 0.4% and Count Vectorizer achieved an accuracy of 99.18% with a false positive rate of 0.98% respectively for three datasets used.

“A Combine Model for Email Classification in Hindi Language using Supervised Learning (NB, K-NN, DT, SVM)”

Email communication is necessary in today's environment, yet unwanted emails create issue in such communication. The current study focuses on developing an Email classification model for the use of classifiers approaches. The goal of this research is to classification of emails based on features. For classification especially in Hindi language of the email dataset Different machine learning classifiers such as Naïve Bayes, Decision Tree, K-Nearest Neighbor and Support Vector Machine used in research work as well as we used combined model also for optimum accuracy.

“A Combine Model for Email Classification in Hindi Language using Supervised Learning (NB, K-NN, DT, SVM)”

Email communication is necessary in today's environment, yet unwanted emails create issue in such communication. The current study focuses on developing an Email classification model for the use of classifiers approaches. The goal of this research is to classification of emails based on features. For classification especially in Hindi language of the email dataset Different machine learning classifiers such as Naïve Bayes, Decision Tree, K-Nearest Neighbor and Support Vector Machine used in research work as well as we used combined model also for optimum accuracy.

An E-mail Classification Algorithm based on Stacking Integrated Learning

<p>The text filtering of traditional anti spam system mainly focuses on keyword matching and text fingerprint analysis, which is difficult to accurately identify and classify spam. Therefore, an integrated learning algorithm based on stackin g is proposed in this paper. Firstly, the algorithm takes the manually marked text data of various categories as samples, uses TF-IDF algorithm to train the word vector space model, then selects linear SVC, xgboost and logistic regression algorithm to structure the base classifier, uses random forest algorithm to structure the meta classifier, and combines the stacking ensemble learning algorithm to structure the classification model. It achieves the function of dividing e-mail into five categories: illegal, advertisement, news, bill and recruitment. From the simulation results, the AUC values of the stacking integrated learning classification algorithm for each category are 0.92, 0.95, 1.00, 0.93 and 0.97 respectively, and the AP values are 0.86, 0.88, 1.00, 0.88 and 0.94 respectively, which realizes the high performance and high precision of text classification.</p> <p> </p>

A Study on Various Phishing Techniques and Recent Phishing Attacks

Now-a-days internet has become a very unsafe space to deal with. Hackers are constantly trying to gain the user's personal information, and detailed credentials. So many websites on the internet, even though safe, this safety cannot be assured by all websites. These rule breakers avoid abiding by rules, and try to employ methods like trickery and hacking to gain illegal access to private information. To be able to overcome this problem, we need to first understand the intricacies of how the virus is designed. This paper mainly deals with the different phishing techniques and recent phishing attacks that took place during COVID 19. like Link Manipulation, Filter Evasion, Website Forgery, Phone Phishing and Website Forgery. We have also studied a subtle method to perform phishing attacks that makes links appear legitimate, but actually redirect a victim to an attackers website called Convert Redirect. In this paper, we present some phishing examples like Paypal phishing which involves sending an email that fraudulently claims to be from a well known company and Rapidshare Phishing where in the spoofed web page, phishers attempt to confuse their victims just enough to entice them to enter their login name and password. To perform these types of phishing the Phishers uses so many phishing techniques like Link Manipulation, Filter Evasion, Website Forgery, Phone Phishing and Website Forgery. Phishing techniques include the domain of email messages. Phishing emails have hosted such a phishing website, where a click on the URL or the malware code as executing some actions to perform is socially engineered messages. Lexically analyzing the URLs can enhance the performance and help to differentiate between the original email and the phishing URL. As assessed in this study, in addition to textual analysis of phishing URL, email classification is successful and results in a highly precise anti phishing. From the thorough analysis of the research paper, we have understood how phishing attacks work and the different methods employed to carry out the attack. Also, we have studied some of the most recent phishing attacks and measures taken by the authorities to overcome and prevent any such attacks in future.

Enhancement of email spam detection using improved deep learning algorithms for cyber security

Email has sustained to be an essential part of our lives and as a means for better communication on the internet. The challenge pertains to the spam emails residing a large amount of space and bandwidth. The defect of state-of-the-art spam filtering methods like misclassification of genuine emails as spam (false positives) is the rising challenge to the internet world. Depending on the classification techniques, literature provides various algorithms for the classification of email spam. This paper tactics to develop a novel spam detection model for improved cybersecurity. The proposed model involves several phases like dataset acquisition, feature extraction, optimal feature selection, and detection. Initially, the benchmark dataset of email is collected that involves both text and image datasets. Next, the feature extraction is performed using two sets of features like text features and visual features. In the text features, Term Frequency-Inverse Document Frequency (TF-IDF) is extracted. For the visual features, color correlogram and Gray-Level Co-occurrence Matrix (GLCM) are determined. Since the length of the extracted feature vector seems to the long, the optimal feature selection process is done. The optimal feature selection is performed by a new meta-heuristic algorithm called Fitness Oriented Levy Improvement-based Dragonfly Algorithm (FLI-DA). Once the optimal features are selected, the detection is performed by the hybrid learning technique that is composed of two deep learning approaches named Recurrent Neural Network (RNN) and Convolutional Neural Network (CNN). For improving the performance of existing deep learning approaches, the number of hidden neurons of RNN and CNN is optimized by the same FLI-DA. Finally, the optimized hybrid learning technique having CNN and RNN classifies the data into spam and ham. The experimental outcomes show the ability of the proposed method to perform the spam email classification based on improved deep learning.

Is the key to phishing training persistence?: Developing a novel persistent intervention.

Most previous phishing interventions have employed discrete training approaches, such as brief instructions aimed at improving phishing detection. However, these discrete interventions have demonstrated limited success. The present studies focused on developing an alternative to discrete training by providing college-age adults with a persistent classification aid that guided them on what characteristics a phishing email might include. Experiment 1 determined if this classification aid improved email categorization performance relative to feedback and control. Experiment 2 continued the evaluation of the classification aid to determine whether performance improvements were due to increased systematic processing of emails. Experiment 3 explored whether the classification aid would be more effective when embedded directly into the email interface. The results suggested three major findings. (a) Persistent interventions can improve phishing email detection. (b) Performance improvements were largest when the classification aid was embedded into the task. (c) These benefits were likely driven by an improved systematic processing of the emails. This novel phishing classification aid serves as a promising persistent intervention that can be adaptable to specific email environments and individuals. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

Exploring the effectiveness of word embedding based deep learning model for improving email classification

PurposeClassifying emails as ham or spam based on their content is essential. Determining the semantic and syntactic meaning of words and putting them in a high-dimensional feature vector form for processing is the most difficult challenge in email categorization. The purpose of this paper is to examine the effectiveness of the pre-trained embedding model for the classification of emails using deep learning classifiers such as the long short-term memory (LSTM) model and convolutional neural network (CNN) model.Design/methodology/approachIn this paper, global vectors (GloVe) and Bidirectional Encoder Representations Transformers (BERT) pre-trained word embedding are used to identify relationships between words, which helps to classify emails into their relevant categories using machine learning and deep learning models. Two benchmark datasets, SpamAssassin and Enron, are used in the experimentation.FindingsIn the first set of experiments, machine learning classifiers, the support vector machine (SVM) model, perform better than other machine learning methodologies. The second set of experiments compares the deep learning model performance without embedding, GloVe and BERT embedding. The experiments show that GloVe embedding can be helpful for faster execution with better performance on large-sized datasets.Originality/valueThe experiment reveals that the CNN model with GloVe embedding gives slightly better accuracy than the model with BERT embedding and traditional machine learning algorithms to classify an email as ham or spam. It is concluded that the word embedding models improve email classifiers accuracy.