Email Spam Classification Research Articles

Exploring Machine Learning Algorithms for Email Spam Filtering

Spams are irrelevant or inappropriate emails, and they are on the rise across different digital communication channels globally. These unsolicited and unwelcome perilous emails lead to adverse impacts like security risks, time and resources wastage, financial loss, reputation damage, and legal consequences. Therefore, intelligent anti-spam email filters are required to prevent such emails from being delivered to the recipient's mailbox. Artificial Intelligence and Machine Learning algorithms provide trustworthy countermeasures and affordable solutions to tackle this global problem. Machine Learning techniques like Logistics Regression, Decision Tree, Support Vector Machines, K-nearest neighbor, Naïve Bayes Classifier, and Random Forest provide solutions to classify spam emails with reasonable accuracy and precision. This research explores different approaches to identifying spam emails and developing machine-learning algorithms to filter such emails automatically. Results manifest that the Support Vector Machine and Naïve Bayes Model most accurately classified spam emails. The Support Vector Machine exhibited the best balance between high precision and recall, making it highly effective in identifying spam messages with fewer false positives.

Next-Generation Spam Filtering: Comparative Fine-Tuning of LLMs, NLPs, and CNN Models for Email Spam Classification

Spam emails and phishing attacks continue to pose significant challenges to email users worldwide, necessitating advanced techniques for their efficient detection and classification. In this paper, we address the persistent challenges of spam emails and phishing attacks by introducing a cutting-edge approach to email filtering. Our methodology revolves around harnessing the capabilities of advanced language models, particularly the state-of-the-art GPT-4 Large Language Model (LLM), along with BERT and RoBERTa Natural Language Processing (NLP) models. Through meticulous fine-tuning tailored for spam classification tasks, we aim to surpass the limitations of traditional spam detection systems, such as Convolutional Neural Networks (CNNs). Through an extensive literature review, experimentation, and evaluation, we demonstrate the effectiveness of our approach in accurately identifying spam and phishing emails while minimizing false positives. Our methodology showcases the potential of fine-tuning LLMs for specialized tasks like spam classification, offering enhanced protection against evolving spam and phishing attacks. This research contributes to the advancement of spam filtering techniques and lays the groundwork for robust email security systems in the face of increasingly sophisticated threats.

Machine Learning Based Classification for Spam Detection

Electronic Electronic messages, i.e. e-mails, are a communication tool frequently used by individuals or organizations. While e-mail is extremely practical to use, it is necessary to consider its vulnerabilities. Spam e-mails are unsolicited messages created to promote a product or service, often sent frequently. It is very important to classify incoming e-mails in order to protect against malware that can be transmitted via e-mail and to reduce possible unwanted consequences. Spam email classification is the process of identifying and distinguishing spam emails from legitimate emails. This classification can be done through various methods such as keyword filtering, machine learning algorithms and image recognition. The goal of spam email classification is to prevent unwanted and potentially harmful emails from reaching the user's inbox. In this study, Random Forest (RF), Logistic Regression (LR), Naive Bayes (NB), Support Vector Machine (SVM) and Artificial Neural Network (ANN) algorithms are used to classify spam emails and the results are compared. Algorithms with different approaches were used to determine the best solution for the problem. 5558 spam and non-spam e-mails were analyzed and the performance of the algorithms was reported in terms of accuracy, precision, sensitivity and F1-Score metrics. The most successful result was obtained with the RF algorithm with an accuracy of 98.83%. In this study, high success was achieved by classifying spam emails with machine learning algorithms. In addition, it has been proved by experimental studies that better results are obtained than similar studies in the literature.

Enhancing Spam Email Classification Using Effective Preprocessing Strategies and Optimal Machine Learning Algorithms

Objective: This article proposes a content-based spam email classification by applying various text pre-processing techniques. NLP techniques have been applied to pre-process the content of an email to get the optimal performance of spam email classification using machine learning. Method: Several combinations of pre-processing methods, such as stopping, removing tags, converting to lower case, removing punctuation, removing special characters, and natural language processing, were applied to the extracted content from the email with machine learning algorithms like NB, SVM, and RF to classify an email as ham or spam. The standard datasets like Enron and SpamAssassin, along with the personal email dataset collected from Yahoo Mail, are used to evaluate the performance of the models. Findings: Applying stemming in pre-processing to the RF classifier yielded the best results, achieving 99.2% accuracy on the SpamAssassin dataset and 99.3% accuracy on the Enron dataset. Lemmatization followed closely with 99% accuracy. In real-world testing on a personal Yahoo email dataset, the proposed method significantly improved accuracy from 89.82% to 97.28% compared to the email service provider's built-in classifier. Additionally, the study found that SVM performs accurately when stop words are retained. Novelty: This article introduces a unique perspective by highlighting the fine-tuning of pre-processing techniques. The focus is on removing tags and certain special characters, while retaining those that improve spam email classification accuracy. Unlike prior works that primarily emphasize algorithmic approaches and pre-defined processing functions, our research delves into the intricacies of data preparation, showcasing its significant impact on spam email classifiers. These findings emphasize the crucial role of pre-processing and contribute to a more nuanced understanding of effective strategies for robust spam detection. Keywords: Spam, Classification, Pre-processing, NLP, Machine Learning

Efficient Email Spam Classification with N-gram Features and Ensemble Learning

In this paper, we present an innovative approach to enhancing email spam classification using N-gram features, TF-IDF weighting, SMOTE oversampling, and ensemble learning techniques such as Decision Trees, Random Forests, and Ensemble Extra Trees. Our methodology involves preprocessing the dataset to extract N-gram features, applying TF-IDF weighting to highlight important terms, and addressing class imbalance through SMOTE. We then train and evaluate multiple classification models and find that the Ensemble Extra Trees algorithm outperforms others in terms of accuracy, precision, recall, and F1-score. Our experiments on benchmark datasets confirm the efficacy of our approach, showcasing significant improvements in spam detection accuracy and highlighting the potential of ensemble learning for email spam classification. This research contributes to the advancement of spam filtering technologies, providing a robust and efficient solution for accurately identifying and categorizing spam emails.

Spam email classification based on SVM, Transformer and Naive Bayes

As a matter of fact, with the booming of information and big data, there are too many unwanted e-mails called spam sent to peoples e-mail account in recent years. On this basis, it could lead to a lot of problems including occupying the public resources, causing financial loss and so on. With this in mind, spam filtering technique is in need to solve the problem and address the issues. In reality, based on previous analysis, machine learning methods are very effective in spam filtering. On this basis, this study carried out background research of machine learning algorithms in spam filtering, and find the spam e-mail dataset of Kaggle.com, and implement 3 algorithms on the dataset. According to the analysis, Transformer and SVM work better on the dataset, and SVM is the best. At the same time, the current limitations are discussed as well. In addition, the prospects are demonstrated in the meantime.

Random Forest Machine Learning for Spam Email Classification

This research discusses the crucial role of email as a main element in digital communication, facilitating information transfer and serving as an advertising platform. However, the problem of email spam, which involves sending unsolicited commercial messages, has had negative impacts such as consuming large amounts of resources and disrupting user experience. With its affordable cost and ease of sending messages to thousands of recipients, email spam includes product promotions, pornographic material, viruses and irrelevant content. The impact includes loss of time and damage to the user's computer resources. To address this problem, email services provide advanced spam filters that use email content analysis and machine learning techniques. This research focuses on the use of the Random Forest Classification algorithm as a basis for filtering spam emails. Although Random Forest is known to have strong classification capabilities, the risk of overfitting is a challenge. Therefore, this study adopts the Randomized Search CV method to identify the best parameter combination, ensuring the reliability of the model in dealing with the complexity of diverse email datasets. With this approach, this research contributes to the development of effective solutions to reduce the impact of email spam in digital communications.

Detecting Spam E-mails with Content and Weight-based Binomial Logistic Model

Spam e-mails are continuously increasing and are a serious threats to a network and its users. Several efficient methods are available regarding this context, but still, it is evolving randomly. Considering this, the proposed approach addresses the problem of spam detection by combining traditional content-matching criteria with the modified version of the binomial logistic algorithm. The work generates seven categories for content-matching, which begins from three basic categories, namely: special words, adult content, and specific symbols and digits. The remaining four categories are derived from various possible combinations of these basic categories. The words selected for each category are carefully curated based on the human psychology of action and reaction. Then, a weight is assigned to each of the categories to signify their importance and a threshold criterion is deployed before implementing the binomial logistic algorithm, which not only increases the efficiency of the proposed algorithm but also reduces the rate of misclassification. The proposed model is tested on six separate datasets of Enron Spam Corpus, where 98.31% and 92.575% are the maximum and minimum accuracies achieved, respectively, in spam e-mail classification. The AUC_ROC scores for the entire Spam Corpus range between 0.927 and 0.983. A comparison is also carried out between the proposed algorithm and the other methods of spam detection that have logistic regression. Finally, the suggested method can adequately handle a large sample size without compromising the efficacy, which is measured using accuracy, precision, recall, F-measure, and AUC_ROC score.

Diverse ensemble classifier driven Email spam classification using multiple word embedding’s with COCOB optimizer

Spam Email is a serious concern which can steal user’s personal information and cause huge financial loss due to the increasing rate of internet users. Therefore, the demand for accurate spam filtering has become more sophisticated for the Email spam detection. In the existing techniques, it is difficult to intricate the relationship between words in the Email using certain word embedding techniques and learning rate tuning is one of the greatest challenges of stochastic optimization. To overcome this difficulty, the proposed framework uses diverse ensemble based Email spam classification by incorporating multiple word embedding’s with Continuous Coin Betting optimizer. Word2Vec is used to produce the first set of 200D, next set of 200D word embedding is produced by Glove and 768D is produced by using Bidirectional Encoder Representations from Transformers (BERT) respectively. After generating word embedding, then it is classified through diverse ensemble based classifier with base level classifier consists of Long Short Term Memory (LSTM) Networks, Gated Recurrent Unit (GRU) and Bi-directional Gated Recurrent Unit (Bi-GRU) and LSTM as Meta-classifier using COCOB optimizer. Experiments were conducted on 3 benchmark Email dataset and result shows that the proposed system outperforms well with a low false positive rate.

A hybrid correlation-based deep learning model for email spam classification using fuzzy inference system

Spam emails are unwanted and unsolicited messages. The major problems in email spam detection methods are low detection rates and a high likelihood of false alarms. This study proposes a hybrid correlation-based deep learning model for email spam classification using a fuzzy inference system. Using a rule-based hybrid feature selection technique, we choose the most crucial features from a preprocessed spam-based dataset. The selected features are then loaded into a deep learning model for spam classification, and fuzzy logic is employed to categorize each spam class into its severity levels to decrease misclassification. Compared with other machine learning methods, the proposed method shows better F1-score results for both test set spambase and test set non-spambase of 96.5% and 96.4%, respectively. Similarly, the developed method shows better accuracy, error rate, and processing time of 94.0017%, 5.9983%, and 0.5 s, respectively. The proposed approach also shows a reduced misclassification based on the fuzzy inference system.

A Comprehensive Review on Email Spam Classification with Machine Learning Methods

This comprehensive review delves into the realm of email spam classification, scrutinizing the efficacy of various machine learning methods employed in the ongoing battle against unwanted email communication. The paper synthesizes a wide array of research findings, methodologies, and performance metrics to provide a holistic perspective on the evolving landscape of spam detection. Emphasizing the pivotal role of machine learning in addressing the dynamic nature of spam, the review explores the strengths and limitations of popular algorithms such as Naive Bayes, Support Vector Machines, and neural networks. Additionally, it examines feature engineering, dataset characteristics, and evolving threats, offering insights into the challenges and opportunities within the field. With a focus on recent advancements and emerging trends, this review aims to guide researchers, practitioners, and developers in the ongoing pursuit of robust and adaptive email spam classification systems.

Improving spam email classification accuracy using ensemble techniques: a stacking approach

AbstractSpam emails pose a substantial cybersecurity danger, necessitating accurate classification to reduce unwanted messages and mitigate risks. This study focuses on enhancing spam email classification accuracy using stacking ensemble machine learning techniques. We trained and tested five classifiers: logistic regression, decision tree, K-nearest neighbors (KNN), Gaussian naive Bayes and AdaBoost. To address overfitting, two distinct datasets of spam emails were aggregated and balanced. Evaluating individual classifiers based on recall, precision and F1 score metrics revealed AdaBoost as the top performer. Considering evolving spam technology and new message types challenging traditional approaches, we propose a stacking method. By combining predictions from multiple base models, the stacking method aims to improve classification accuracy. The results demonstrate superior performance of the stacking method with the highest accuracy (98.8%), recall (98.8%) and F1 score (98.9%) among tested methods. Additional experiments validated our approach by varying dataset sizes and testing different classifier combinations. Our study presents an innovative combination of classifiers that significantly improves accuracy, contributing to the growing body of research on stacking techniques. Moreover, we compare classifier performances using a unique combination of two datasets, highlighting the potential of ensemble techniques, specifically stacking, in enhancing spam email classification accuracy. The implications extend beyond spam classification systems, offering insights applicable to other classification tasks. Continued research on emerging spam techniques is vital to ensure long-term effectiveness.

Spam email detection using a novel multilayer classification-based decision technique

Because of the rapid advancement of technology over the last several years, the number of internet users is growing at an exponential rate, and as a result, email communication has become popular as a means of exchanging information over the internet. Sending data and communicating with peers via email is the most cost-effective method. These email services also cause problems for users by sending electronic junk mail, often known as spam mail. Spam email is a privacy concern that is linked to a slew of commercial and dangerous websites, causing phishing, virus distribution, and a slew of other problems. This study examines several aspects that have been used for email spam classification, as well as offering an overview of a handful of classifiers or algorithms that have been successfully evaluated, as well as exploratory data analysis. The proposed email spam classifier uses three parallel layers of machine learning and deep learning techniques, followed by a decision function to determine whether or not the emails are spam. During testing, it was found that the proposed classifier beats similar systems on the standard dataset with an accuracy of 98.4%.

Classification of Spam E-mail based on Naïve Bayes Classification Model

With the rising number of spam email, the need of more sufficient antispam filter is surging. Phishing attack can lead to extremely large losses of companies and individual, even more than 1 billion dollars in one year. This paper investigates and combines Naïve Bayes Classification and clustering algorithm in the application of identifying spam emails. With sample emails to create a dynamic dictionary containing most frequent words in spam and normal emails, this distribution of spam filter will provide a stricter method to prevent spam emails than those methods used in mail companies, e.g., Google, Yahoo, and Outlook.com. Besides, this paper also compares several algorithms used today in classifying spams and the future techniques of deep learning and machine learning’s application in classifying spam emails. According to the analysis, Google’s algorithm has the most comprehensive function, but such algorithm has less strict rule than Yahoo’s. Outlook.com, as a combination of Microsoft application, it has a unique algorithm for encrypting and filtering spams. Overall, these results shed light on guiding further exploration of both comprehensive and strict rule for classifying spams.

Classifying spam emails using agglomerative hierarchical clustering and a topic-based approach

Spam emails are unsolicited, annoying and sometimes harmful messages which may contain malware, phishing or hoaxes. Unlike most studies that address the design of efficient anti-spam filters, we approach the spam email problem from a different and novel perspective. Focusing on the needs of cybersecurity units, we follow a topic-based approach for addressing the classification of spam email into multiple categories. We propose SPEMC-15K-E and SPEMC-15K-S, two novel datasets with approximately 15K emails each in English and Spanish, respectively, and we label them using agglomerative hierarchical clustering into 11 classes. We evaluate 16 pipelines, combining four text representation techniques -Term Frequency-Inverse Document Frequency (TF-IDF), Bag of Words, Word2Vec and BERT- and four classifiers: Support Vector Machine, Näive Bayes, Random Forest and Logistic Regression. Experimental results show that the highest performance is achieved with TF-IDF and LR for the English dataset, with a F1 score of 0.953 and an accuracy of 94.6%, and while for the Spanish dataset, TF-IDF with NB yields a F1 score of 0.945 and 98.5% accuracy. Regarding the processing time, TF-IDF with LR leads to the fastest classification, processing an English and Spanish spam email in 2ms and 2.2ms on average, respectively.

E-mail Spam Classification using KNN and Naive Bayes

E-mail spam filtering is becoming a critical and concerned issue in network security recently, and multiple machine learning techniques have been applied to tackle such sort of classification problem. With the emerging of machine learning framework, most of the tasks has been changed via the effective machine learning algorithms with satisfying performance and high speed. However, the underlying performances of different algorithms under certain given circumstances still lack of an intuitive demonstration. Hence, this study mainly focuses on the performance of two widely-used algorithms (KNN and Naive Bayes) from metrics including accuracy and running time, comparing the unique advantage of each algorithm when classifying emails. The paper uses thousands of spam data to feed two algorithms and analyzes both results respectively, indicating that KNN classifier performs better when determining the spam messages while the opposite is true for the Naive Bayes classifier. Thus, designers can pick an appropriate algorithm easily when dealing with spam filter issues under a given dataset whose features and properties are known.

Semantic Graph Based Convolutional Neural Network for Spam e-mail Classification in Cybercrime Applications

Spam is characterized as unnecessary and garbage E-mails. Due to the increasing of unsolicited E-mails, it is becoming more and more crucial for mail users to utilize a trustworthy spam E-mail filter. The shortcomings of spam classifier are defined by their increasing inability to manage large amounts of relevant messages and to effectively detect and effectively detect spam messages. Numerous characteristics in spam classifications are problematic. Given that selecting features is one of the most often used and successful techniques for feature reduction, it is a crucial duty in the identification of keyword content. As a result, features that are unnecessary and pointless yet potentially harm effciency would be removed. In this study, we present SGNNCNN (Semantic Graph Neural Network With CNN) as a solution to tackle the diffcult task of mail identification. By projections E-mails onto a graph and by using the SGNN-CNN model for classifications, this technique transforms the E-mail classification issue into a graph classification challenge. There is no need to integrate the word into a representation since the E-mail characteristics are produced from the semantic network. On several open databases, the technique's effectiveness is evaluated. Some few public databases were used in experiments to demonstrate the high accuracy of the proposed approach for classifying E-mails. In term of spam classification, the performance is superior to state-of-the-art deep learning-based methods.

Hyperparameter Optimization of Ensemble Models for Spam Email Detection

Unsolicited emails, popularly referred to as spam, have remained one of the biggest threats to cybersecurity globally. More than half of the emails sent in 2021 were spam, resulting in huge financial losses. The tenacity and perpetual presence of the adversary, the spammer, has necessitated the need for improved efforts at filtering spam. This study, therefore, developed baseline models of random forest and extreme gradient boost (XGBoost) ensemble algorithms for the detection and classification of spam emails using the Enron1 dataset. The developed ensemble models were then optimized using the grid-search cross-validation technique to search the hyperparameter space for optimal hyperparameter values. The performance of the baseline (un-tuned) and the tuned models of both algorithms were evaluated and compared. The impact of hyperparameter tuning on both models was also examined. The findings of the experimental study revealed that the hyperparameter tuning improved the performance of both models when compared with the baseline models. The tuned RF and XGBoost models achieved an accuracy of 97.78% and 98.09%, a sensitivity of 98.44% and 98.84%, and an F1 score of 97.85% and 98.16%, respectively. The XGBoost model outperformed the random forest model. The developed XGBoost model is effective and efficient for spam email detection.

Spam email classification and sentiment analysis based on semantic similarity methods

Spam email classification and sentiment analysis based on semantic similarity methods

Spam email classification and sentiment analysis based on semantic similarity methods

Spam email classification and sentiment analysis based on semantic similarity methods