Spam Emails Research Articles

Email spam detection using bagging and boosting of machine learning classifiers

Email spam detection using bagging and boosting of machine learning classifiers

Feature Extraction and Classification of Email Spam Detection Using IMTF-IDF+Skip-Thought Vectors

Spam is a major concern in present emails, and there are several reasons for sending spam emails. The two most common ones are advertising and fraud. If supported by suitable preprocessing approaches, the detection algorithm for spam email or spam classifier will function effectively (removal of noise, removal of stop words, stemming, lemmatization, term frequency). Spam that combines both text and image components is referred to as hybrid spam. Compared to spam emails with images and text, it is more unsafe and complex. To distinguish spam or ham, we must use an effective and smart approach in order to have a strong representation of emails and improve classification performance. In this paper, we propose a multi-modal architecture relying on a feature model (MMA-FM) that concatenates two embedding vectors. The text and image sections of the similar emails were separated using a hybrid model (IMTF-IDF+Skip-thoughts) and the convolutional neural network (CNN) as a feature extraction technique. The extracted features are concatenated and given to Naïve Bayes (NB) and Support Vector Machine (SVM) models to classify hybrid email as either spam or ham. In this paper we used two hybrid datasets: Enron, Dredze, and TREC 2007, which are publicly accessible corpora. Our results show that the SVM model provides an accuracy of 99.16%, which is higher when compared to the Naïve Bayes method.

Detection of Spam Email

Spam, often known as unsolicited email, has grown to be a major worry for every email user. Nowadays, it is quite challenging to filter spam emails since they are made, created, or written in such a unique way that anti-spam filters cannot recognize them. In order to predict or categorize emails as spam, this paper compares and reviews the performance metrics of a few categories of supervised machine learning techniques, including Svm (Support Vector Machine), Random Forest, Decision Tree, Cnn, (Convolutional Neural Network), Knn(K Nearest Neighbor), Mlp(Multi-Layer Perceptron), Adaboost (AdaptiveBoosting), and Nave Bayes algorithm. Thegoal of this study is to analyze the specificsor content of the emails, discover a limited dataset, and create a classification model that can predict or categorize whether spam is present in an email. Transformers’ Bidirectional Encoder Representations) has been optimized to perform the duty of separating spam emails from legitimate emails (Ham). To put the text’s context into perspective, Bert uses attention layers. Results are contrasted with a baseline Dnn (deep neural network) modelthat consists of two stacked Dense layers and a Bilstm (bidirectional Long Short-Term Memory) layer. Results are also contrasted with a group of traditional classifiers, including k- Nn (k-nearest neighbours) and Nb (Naive Bayes). The model is tested for robustness andpersistence using two open-source data sets, one of which is utilized to train the model.

Artificial Intelligence Model for Email Spam Detection

Artificial Intelligence Model for Email Spam Detection

Are we Cyber aware? A cross sectional study on the prevailing Cyber practices among adults from Thiruvalla, Kerala.

Background: Cyber-crime is described as any unlawful activity which is committed using any computing devices, like computer/smartphone and which is a part of internet. There are different methods by which cyber-crime is committed which include 1) attacks on computer systems, 2) cyber-bullying, 3) email spam, 4) phishing, 5) identity theft. Breaches in cyber security have become a severe danger to world security and the economy, compromising essential infrastructure and wreaking havoc on company performance, resulting in significant cognitive property loss. It is a sad reality that cybercrime cases have witnessed a steady spike. India has witnessed a significant increase in cases of cyber fraud and various cyber-related incidents in the last three years. The present study was done to assess the prevailing cyber practices among adults from Thiruvalla, Kerala.Methods: The present cross-sectional study was conducted among 340 adults in Thiruvalla, Kerala from January to June, and 2022. A semi-structured questionnaire was used to elicit information from the study participants after obtaining consent. The quantitative data collected was analysed using the software statistical package for social sciences. The results have been presented as tables and charts showing frequencies and percentages.Results: 133 out of the 350 study participants (38%) believed that it was important to be aware of cyber security risks in general while 52% (182) of the study participants considered cyber security awareness to be the only solution to the existing online scams.Conclusions: Widespread awareness campaigns are necessary to improve the cyber awareness of the community and thereby improve their cyber practices.

Application of Artificial Intelligence to improve Customer Understanding: Transformer based Topic Modeling in practice

Since the past few years, so-called pre-trained Language Models (PLM) are considered state-of-the-art in the field of Natural Language Processing (NLP) and are thus experiencing widespread and successful application. In addition to traditional supervised Machine Learning (ML) tasks such as spam email or customer churn classification, this technology opens up advanced approaches to unsupervised learning and data analytics in general. One of particular interest is the automatic identification of latent topics within a large collection of texts, also known as Topic Modeling (TM). Such modelling approaches offer great potential, especially for industrial environments as well as the consumer goods market, to explore increasing amounts of data from diverse and constantly growing data sources. As a holistic concept, it can be utilized in a highly targeted and efficient manner for applications through an optimized combination of Artificial Intelligence (AI) and Cloud Computing (CC) systems.

Detecting Spam Email With Machine Learning Optimized With Bio-Inspired Metaheuristic Algorithms

Electronic mail has eased communication methods for many organizations as well as individuals. Spammers use this strategy to make fraudulent gains by sending unsolicited emails. This project aims to present a method for detection of spam emails with machine learning algorithms that are optimized with bio-inspired methods. A literature review is carried to explore the efficient methods applied on different datasets to achieve good results. The bio-inspired algorithms like Particle Swarm Optimization and Genetic Algorithm were implemented to optimize the performance of classifiers. Multinomial Naïve Bayes with Genetic Algorithm performed the best overall. The comparison of our results with other machine learning and bio-inspired models to show the best suitable model is also discussed.

Deliver

Deliver

SubStop: An analysis on subscription email bombing attack and machine learning based mitigation

Email Bombing, a kind of denial-of-service (DoS) attack is crippling internet users and is on the rise recently. A particularly notorious type is the Subscription Bombing attack, where a victim user’s inbox is bombarded with a stream of subscription emails at a particular period. This kind of attack helps the perpetrator to hide their real motive in lieu of a barrage of legitimate-looking emails. The main challenge for detecting subscription bombing attacks is that most of the attacking email appears to be legitimate and benign and thus can bypass existing anti-spam filters. In order to shed some light on the direction of detecting the bombing attacks, in this paper we first conduct some reverse engineering study on the Gmail anti-spam mechanism (as the information is not publicly available) and in-depth feature analysis of real-life bombing attack emails. Leveraging the insights from our reverse engineering study and data analysis, we propose a novel layered detection architecture, termed as SubStop, to detect and mitigate subscription bombs. SubStop exploits the statistics of incoming volume, source domain distribution, the correlation among different features, and implements machine learning to achieve effective detection. In specific, we utilize the weighted support vector machine (WSVM) and properly tune the class weights to achieve high accuracy in detecting bombing attacks. Despite the scarcity of public email data sets, we conduct extensive experiments on a real-life subscription bomb attack and real-time attacks using our bombing simulation script (which is facilitated by our reverse engineering findings), on test email accounts. Detailed experimental results show that our proposed architecture is very robust and highly accurate in detecting and mitigating a subscription bombing attack.

Stability and Effective Process Control for Secure Email Filtering

A fantastic tool for both commercial and personal communication is electronic mail. It has increasingly become a necessary component of our working life since it is straightforward, available, and simple to use. Spam emails have started to tarnish internet experiences and threaten the integrity of email. Due to the exponential growth of spam, both people and organisations are under a great deal of financial and other strain. In order to prevent the future of email itself from being in jeopardy, a solution to the spam problem must be discovered. There is an urgent need to solve the Email spam issue since spam volume has been rising over the last several decades. As part of this effort, many effects of spam emails on businesses and people were noted and thoroughly examined. In order to properly assess current technologies, solutions, and methods, a comprehensive literature review was conducted throughout the procedures. The goals of this work is to develop new methodologies for the implementation of new strategies for the efficient management of email spam and to construct a proof-of-concept software system for the Process controlled assessment of such strategies.

Performance Evaluation of LSTM and RNN Models in the Detection of Email Spam Messages

Email spam is an unwanted bulk message that is sent to a recipient’s email address without explicit consent from the recipient. This is usually considered a means of advertising and maximizing profit, especially with the increase in the usage of the internet for social networking, but can also be very frustrating and annoying to the recipients of these messages. Recent research has shown that about 14.7 billion spam messages are sent out every single day of which more than 45% of these messages are promotional sales content that the recipient did not specifically opt-in. This has gotten the attention of many researchers in the area of natural language processing. In this paper, we used the Long Short-Time Memory (LSTM) for classification tasks between spam and Ham messages. The performance of LSTM is compared with that of a Recurrent Neural Network( RNN) which can also be used for a classification task of this nature but suffers from short-time memory and tends to leave out important information from earlier time steps to later ones in terms of prediction. The evaluation of the result shows that LSTM achieved 97% accuracy with both Adams and RMSprop optimizers compared to RNN with an accuracy of 94% with RMSprop and 87% accuracy with Adams optimizer.

Application of Natural Language Processing and Machine Learning Boosted with Swarm Intelligence for Spam Email Filtering

Spam represents a genuine irritation for email users, since it often disturbs them during their work or free time. Machine learning approaches are commonly utilized as the engine of spam detection solutions, as they are efficient and usually exhibit a high degree of classification accuracy. Nevertheless, it sometimes happens that good messages are labeled as spam and, more often, some spam emails enter into the inbox as good ones. This manuscript proposes a novel email spam detection approach by combining machine learning models with an enhanced sine cosine swarm intelligence algorithm to counter the deficiencies of the existing techniques. The introduced novel sine cosine was adopted for training logistic regression and for tuning XGBoost models as part of the hybrid machine learning-metaheuristics framework. The developed framework has been validated on two public high-dimensional spam benchmark datasets (CSDMC2010 and TurkishEmail), and the extensive experiments conducted have shown that the model successfully deals with high-degree data. The comparative analysis with other cutting-edge spam detection models, also based on metaheuristics, has shown that the proposed hybrid method obtains superior performance in terms of accuracy, precision, recall, f1 score, and other relevant classification metrics. Additionally, the empirically established superiority of the proposed method is validated using rigid statistical tests.

Multi-Objective Personalization in Multi-Stakeholder Organizational Bulk E-mail

Bulk email is often used in organizations to communicate "important-to-organization'' messages such as policy changes, organizational plans, and administrative updates. However, normal employees may prefer messages more relevant to their jobs or interests. Organizations face the challenge of balancing prioritizing the messages they prefer employees to know (tactical goals) while maintaining employees' positive experiences with these bulk emails, then they continue to read these emails in the future (strategic goals). Could personalization help organizations achieve these tactical and strategic goals? In an 8-week field experiment with a university newsletter, we implemented a 4x5x5 factorial design on personalizing subject lines, top news, and message order based on both the employees' and the organization's preferences. We measured these designs' influences on the open/interest/recognition/read-in-detail rate of the whole newsletter and the single messages within it. We found that ''important-to-organization'' messages only got higher recognition rates when being put on subject lines / top news (tactical goal). Mixing them with employee-preferred messages in top news did not bring further improvement to their own recognition rates but could improve the whole newsletter's recognition rate. Only when the top news solely contained the employee-preferred messages were the employees slightly more interested in the newsletter (strategic goal). We further analyze on which topics the employees and the organization's preferences conflicted. Finally, we discuss the design suggestions for organizational bulk email.

Hierarchical Dirichlet and Pitman–Yor process mixtures of shifted‐scaled Dirichlet distributions for proportional data modeling

AbstractIn this article, first, we propose a novel unsupervised learning method based on a hierarchical Dirichlet process mixture of shifted‐scaled Dirichlet (SSD) distributions. Second, we extend it to a hierarchical Pitman–Yor process mixture of SSD distributions. The goal is to find a model that properly fits complex real‐world data. Our models are based on SSD distributions that are more flexible than Dirichlet distribution in fitting proportional data. Simultaneous data fitting (parameter estimate) and model selection (model complexity determination) are possible with the suggested methods. We applied batch and online variational inference for learning the models. The online setting allows us to feed our models with large‐scale streaming data. The effectiveness of our proposed models is evaluated by four realistic and challenging applications, namely, spam email detection, texture clustering, traffic sign detection, and vehicle detection. Experimental results demonstrate the potential of our models to fit proportional data.

Predatory conferences: a systematic scoping review

ObjectiveTo systematically map the scholarly literature on predatory conferences and describe the present state of research and the prevalent attitudes about these conferences.MethodsThis scoping review follows Preferred Reporting Items for...

Medical Resident Awareness of Predatory Journal Practices in an International Medical Education System

ABSTRACT Introduction Learning research methodology is increasingly becoming an essential part of graduate medical education worldwide, with many regulatory and accreditation bodies requiring residents to participate in scholarship. Research methodology workshops have become a standard part of medical curricula; however, there is limited data on how much training on journal selection and the publication process trainees receive. The alarming growth of predatory journals has made it increasingly difficult for researchers, especially trainees and early career physicians, to distinguish these publications from reputable journals. The purpose of this study is to assess the knowledge of reputable and predatory publishing practices amongst medical trainees in an international medical education setting in the United Arab Emirates. Methods A survey on credible journal practices based on the ‘Think. Check. Submit’ initiative was sent to all graduate medical education trainees at two large academic medical centers in Abu Dhabi, United Arab Emirates. Descriptive statistics were used to tabulate variable frequencies. Results Over half of the 160 respondents reported receiving prior research methodology training and 42.5% had at least one publication. The majority of the trainees selected impact factor and the quality of the peer-review process as characteristics of reputable journals. Ambiguous editorial board and rapid publication process were recognized as characteristics of predatory journals by >65% of trainees, however, 95% of all trainees were unaware of Beall’s list or other resources to help select a journal for publication. 15.2% of trainees who received unsolicited emails from publishers submitted their manuscripts to the unfamiliar journals, citing peer recommendation and pressure to publish from their training programs as reasons. Conclusion Trainees in the United Arab Emirates were mostly unaware of reputable publication practices and are vulnerable to publishing in predatory journals. Policy and educational reform are necessary to maintain the credibility and integrity of the scientific process.

Academic spam mails from “Predatory Journals” in cardiology: a cross-sectional study over 3 months

Abstract Background Numerous online journals are referred to as predatory journals because their efforts are focused primarily on collecting publication fees rather than on content, peer review, or manuscript presentation. Many of these journals send Academic Spam Mails (ASM), requesting manuscript submissions. Purpose/Methods Over a 3-month period, all ASMs addressed to the author were collected and analyzed. A fake publication was created with a commonly used structure (Introduction, Methods, Results and Discussion). The article just repeated the phrase “Please get me off your mailing list” over and over again. This publication was submitted in response to each request to test the control mechanisms of the respective journals. Results During the study period, excluding duplicate requests, 125 ASMs (average 2.1 per business day) were received from a wide variety of journals spanning 38 different subject areas, of which 29/38 (76.4%) were from the medical field. 16/125 (16%) were related to a previously published article but not among the author's most important publications (median h-index rank 22; range 10–138). The majority (73.6%) of journals were listed in Beall's predatory journal list, only an absolute minority (0.8%) were listed in the Directory of Open Access Journals, and 96% were not listed in PubMed. Original papers (94.4%), reviews (88.0%), and case reports (87.2%) were most frequently requested; the publication charge (APC) was a median of $1,199 (range $50–3,586); in about 22% of cases, it was not apparent whether APCs were charged and, if so, how much. For the 125 nonsense publications sent out, there was a response in 16/125 (13%) (6 acknowledgements of receipt, 3 apologies, 2 the journal comments that the publication consisted of only one sentence, and 5 other responses). Interestingly, in one case editorial board members had also created a nonsense publication titled: “Ok we not send mail to you” and sent it to the author. 21.6% of the journals had sent ASM to the author again during the next months. Conclusions Academic spam mails from “predatory” online journals are a common problem. They are recognizable by their appearance in warning lists or non-appearance in directories of reputable open access journals. A placement of a nonsense publication in such a “predatory journal” was not successful. Funding Acknowledgement Type of funding sources: None.

Email Spam Detection using Naïve Bayes Algorithm

Abstract: Email Spam has become a vital issue currently, with high-speed growth of internet users. Some people are using them for illegal conducts, phishing and fraud. Sending malicious link through spam emails which can harm our system and may also they will seek into our system. The need of email spam detection is to prevent spam messages from lagging into user’s inbox so it’ll improve user experience. This project will identify those spam emails by using machine learning approach. Machine learning is one amongst the applications of Artificial Intelligence that allow systems to read and improve from experience without being specific programmed. This paper will discuss the machine learning algorithm which is Naïve Bayes. It is a probabilistic classifier, which means it predicts on the idea of the probability of an object and it is selected for the email spam detection having best precision and accuracy.

Trigonometric words ranking model for spam message classification

AbstractThe significant increase in the volume of fake (spam) messages has led to an urgent need to develop and implement a robust anti‐spam method. Several of the current anti‐spam systems depend mainly on the word order of the message in determining the spam message, which results in the system's inability to predict the correct type of message when the word order changes. In this paper, a new framework is proposed for anti‐spam filtering that does not depend on the word's position in the message, called the Trigonometric Words Ranking Model (TWRM). The proposed TWRM is based on restricting spammers over the network by measuring a theta angle, which is a relationship between message weight and spam. TWRM classifies messages by calculating the rank of each word that places the corresponding message in the correct class. The rank of words is derived from their frequency in the entire data category. The proposed method is applied to three datasets of spam messages: UCI spam email, Enron spam, and TREC spam data. The proposed model is proven as more efficient than the Minhash and vector space models. Moreover, the TWRM performance provided better retrieval time and defence, which is reflected in the accuracy of (99.64%), which is higher than that of Minhash (88.79%) and vector space (92.59%).

A Review of Artificial Intelligence Applications in Machine Learning in Mordren World

The newest and most fascinating technique in artificial intelligence is machine learning. Many of the applications we use employ everyday learning techniques. Because Google or Microsoft have incorporated a learning mechanism by which online pages are ranked, they operate effectively as search engines. This learning also recognizes the photographs of pals while browsing Facebook. The learning system also aids in recognizing spam emails, saving the user the hassle of having to waste time sorting through the volume of spam emails. Therefore, this study has explored how machine learning functions as well as how it might be useful in the future.