Academia.eduAcademia.edu

Outline

Title
Abstract
Figures
Introduction
Data Mining
Purpose of the Study
Literature Review
Evaluation of Results
Conclusion
References
All Topics
Computer Science
Machine Learning

SPAM EMAIL DETECTION USING MACHINE LEARNING INTEGRATED IN CLOUD

Profile image of Joyece JaneJoyece Jane
https://doi.org/10.1109/ICNWC57852.2023.10127237
visibility

…

description

8 pages

link

1 file

Abstract

In this project, we focus on electronic mail, one of the most important means of communication among information professionals. As its use and significance among the general populace grows, so does its importance and utility. It has allowed for more adaptability and convenience in communication, both in the private and professional spheres. The increased use of email has led to a rise in spam as well as legitimate messages. An email that is sent to a large number of people without the sender's knowledge or consent is considered spam. Millions of internet users, both casual and professional, are currently frustrated by the widespread problem of email spam. The purpose of this study is to provide a hybrid approach to machine learning for identifying spam in email. Bagging and boosting of machine learning-based multinomial Decision Tree, Naive Bayes, KNN, Random Forest, and the SVM method are the proposed hybrid techniques. The bagging method uses a concurrent combination of weak classifiers to boost classification accuracy. The standard deviation of misclassifications is decreased by using bagging. Alternatively, by linking the classifiers in a series fashion, the boosting strategy can construct a robust classifier out of two or more relatively weak classifiers. Improved classification results can be achieved through reduced bias and variance thanks to the use of boosting. In order to detect spam in emails, it is necessary to take into account datasets, pre-process those datasets, extract and pick features, and classify the data. In this study, we evaluate the feasibility of conducting experiments using data from the Ling-Spam Corpus and the CSDMC2010 Spam Corpus. According to the stop-word list and lemmatiser, Ling-Spam Corpus's dataset is split into four different directories: bare, lemm, lemm stop, and stop. In addition, pre-processing consists of converting strings to word vectors (tokenization), stemming words, and removing stop words. Since the Ling Spam Corpus is already organised according to the stop-word list and the lemmatiser, only the CSDMC2010 Spam Corpus undergoes the stemming and XXX-X-XXXX-XXXX-X/XX/$XX.00 ©20XX IEEE stop words removal processes. Features are extracted and selected from the preprocessed data. The feature selection procedure in this work makes use of a correlation-based approach.

Figures (10)
data patterns [4]. Data mining 1s not a stand-alone strategy; rather, it is an umbrella term for a number of related processes, such as those used in database management, AI, statistics, machine learning, etc. Data mining's main goal is to analyse unprocessed data and produce large amounts of structured information. Figure | depicts a schematic of the data mining/knowledge discovery process [5]. Knowledge, pattern & rules, transformed data, pre-processed data, target data, and raw data are the fundamental components of the data mining process. In order to obtain these elements, a number of steps—from selection and cleaning to pre- processing, feature processing, algorithmic processing, interpretation and evaluation—must take place. Fig 1: Flowchart of the Knowledge Acquisition and Data Mining Procedure Oe ————KO——y———————= ee Email spam is not a new phenomenon; in 1978, Gary Thuerk manually sent the first spam letter on ARPANET to 400 (ARPANET, 2006) persons to attract attention to the launch of his company's DECSYSTEM computer products [6]. The rising incidence of spamming is cause for concern, with 84.85% of all global email correspondence being spam, according to the most recent data from Talos Intelligence (December 2019). [7]. In addition to wasting the recipients’ time, effort, and bandwidth, email spam can lead them to malicious websites that could compromise their computer systems with phishing or malware. Financial scams and terrorist actions can be facilitated by phishing for sensitive information in spam emails. Numerous techniques and filters exist for identifying spam emails. Due to the difficulty of feature selection and the selection of an improper classification strategy, existing methods fail to achieve the necessary performance. This study modifies the bagging and boosting ensemble methods to identify spam in electronic messages. SVM and multinomial naive bayes algorithms are combined as a bagging and boosting technique in machine learning. The ensemble methods can compensate for the weaknesses of the individual ones. The performance of the algorithms used to detect email spam is enhanced by the bagging and data, and raw data are the fundamental components of the
data patterns [4]. Data mining 1s not a stand-alone strategy; rather, it is an umbrella term for a number of related processes, such as those used in database management, AI, statistics, machine learning, etc. Data mining's main goal is to analyse unprocessed data and produce large amounts of structured information. Figure | depicts a schematic of the data mining/knowledge discovery process [5]. Knowledge, pattern & rules, transformed data, pre-processed data, target data, and raw data are the fundamental components of the data mining process. In order to obtain these elements, a number of steps—from selection and cleaning to pre- processing, feature processing, algorithmic processing, interpretation and evaluation—must take place. Fig 1: Flowchart of the Knowledge Acquisition and Data Mining Procedure Oe ————KO——y———————= ee Email spam is not a new phenomenon; in 1978, Gary Thuerk manually sent the first spam letter on ARPANET to 400 (ARPANET, 2006) persons to attract attention to the launch of his company's DECSYSTEM computer products [6]. The rising incidence of spamming is cause for concern, with 84.85% of all global email correspondence being spam, according to the most recent data from Talos Intelligence (December 2019). [7]. In addition to wasting the recipients’ time, effort, and bandwidth, email spam can lead them to malicious websites that could compromise their computer systems with phishing or malware. Financial scams and terrorist actions can be facilitated by phishing for sensitive information in spam emails. Numerous techniques and filters exist for identifying spam emails. Due to the difficulty of feature selection and the selection of an improper classification strategy, existing methods fail to achieve the necessary performance. This study modifies the bagging and boosting ensemble methods to identify spam in electronic messages. SVM and multinomial naive bayes algorithms are combined as a bagging and boosting technique in machine learning. The ensemble methods can compensate for the weaknesses of the individual ones. The performance of the algorithms used to detect email spam is enhanced by the bagging and data, and raw data are the fundamental components of the
3.1. THE DECISION TREE Decision trees have emerged as a popular machine
3.1. THE DECISION TREE Decision trees have emerged as a popular machine
4. EVALUATION OF RESULTS
4. EVALUATION OF RESULTS
Accuracy, F— Measure, F—PR, and TNR.
Accuracy, F— Measure, F—PR, and TNR.
Table 4: FN, FP, TN, and TP Evaluations for the Stoy Category of the Ling Spam Corpus, Based on Machine Learning-based Classifiers Table 5: FN, FP, TN, and TP Values Assessed by Machine Learning-based Classifiers for the CODMC2010 Spam Corpus
Table 4: FN, FP, TN, and TP Evaluations for the Stoy Category of the Ling Spam Corpus, Based on Machine Learning-based Classifiers Table 5: FN, FP, TN, and TP Values Assessed by Machine Learning-based Classifiers for the CODMC2010 Spam Corpus
Table 3: Machine Learning-based classifiers were used to assess the FN, FP, TN, and TP values in the Lemm-Stop nntarnnwm: ntéhlen J tur Onni fLinuniea
Table 3: Machine Learning-based classifiers were used to assess the FN, FP, TN, and TP values in the Lemm-Stop nntarnnwm: ntéhlen J tur Onni fLinuniea
Tables 1 through 5 show that compared to the MNB method, the SVM _ has greater true values (TP and TN) and lower false values (FN and FP). It is necessary to do additional evaluation in terms of performance metrics in order to identify the optimal algorithm. The evaluated values are then utilised in the determination of the Precision, Recall, Accuracy, F- Measure, FNR, FPR, and TNR. Results for the Bare Category in the Ling Spam Corpus as Assessed by Machine Learning-based Classifiers are presented in Table 6. Table 6 shows the results of testing the machine learning classifiers SVM and MNB classifier on the bare category of the ling spam corpus. The SVM outperforms the MNB method in terms of precision, recall, f-measure, accuracy, and TNR, but it performs worse in terms of FNR and FPR. The greater an algorithm's efficacy, the lower its false-measure values (FNR and FPR).
Tables 1 through 5 show that compared to the MNB method, the SVM _ has greater true values (TP and TN) and lower false values (FN and FP). It is necessary to do additional evaluation in terms of performance metrics in order to identify the optimal algorithm. The evaluated values are then utilised in the determination of the Precision, Recall, Accuracy, F- Measure, FNR, FPR, and TNR. Results for the Bare Category in the Ling Spam Corpus as Assessed by Machine Learning-based Classifiers are presented in Table 6. Table 6 shows the results of testing the machine learning classifiers SVM and MNB classifier on the bare category of the ling spam corpus. The SVM outperforms the MNB method in terms of precision, recall, f-measure, accuracy, and TNR, but it performs worse in terms of FNR and FPR. The greater an algorithm's efficacy, the lower its false-measure values (FNR and FPR).

Related papers

Efficient Spam Classification by Appropriate Feature Selection
Manasi Patwardhan

2013

Spam is a key problem in electronic communication, including large-scale email systems and the growing number of blogs. Currently a lot of research work is performed on automatic detection of spam emails using classification techniques such as SVM, NB, MLP, KNN, ID3, J48, RandomTree, etc. For spam dataset it is possible to have large number of training instances. Based on this fact, we have made use of Random Forest and Partial Decision Trees algorithm to classify spam vs non-spam emails. These algorithms outperformed the previously implemented algorithms in terms of accuracy and time complexity. As a preprocessing step we have used feature selection methods such as Chisquare, Information gain, Gain ratio, Symmetrical uncertainty, Relief, OneR and Correlation. This allowed us to select subset of relevant, non redundant and most contributing features to have an added benefit in terms of improvisation in accuracy and reduced time complexity.

View PDFchevron_right
Identifying Potentially Useful Email Header Features for Email Spam Filtering
Ismail Khater

ICDS 2012, The Sixth …, 2012

Email spam continues to be a major problem in the Internet. With the spread of malware combined with the power of botnets, spammers are now able to launch large scale spam campaigns causing major traffic increase and leading to enormous economical loss. In this paper, we identify potentially useful email header features for email spam filtering by analyzing publicly available datasets. Then, we use these features as input to several machine learning-based classifiers and compare their performance in filtering email spam. These classifiers are: C4.5 Decision Tree (DT), Support Vector Machine (SVM), Multilayer Perception (MP), Nave Bays (NB), Bayesian Network (BN), and Random Forest (RF). Experimental studies based on publicly available datasets show that RF classifier has the best performance with an average accuracy, precision, recall, F-Measure, ROC area of 98.5%, 98.4%, 98.5%, and 98.5%, respectively.

View PDFchevron_right
Feature Extraction and Classification of Spam Emails
Nhamo Mtetwa

2018 5th International Conference on Soft Computing & Machine Intelligence (ISCMI)

Emails are a popular and preferred way of written communication in our daily life. The problem with emails is spam. These spam emails are sent with different intentions, but advertisement and fraud are the main reasons. As being inexpensive to send, it causes many problems to the internet society. This paper discusses the use of different feature extraction methods coupled with two different supervised machine learning classifiers evaluated using four performance metrics on two publicly available spam email datasets for spam filtering. We highlight the importance of the correct coupling of feature extraction and classifier, and the merits of using two independent datasets.

View PDFchevron_right
Spam Detection in Email using Machine Learning
R. A. Shehan Sanjula

figshare. Conference contribution., 2022

In today's world, email is used in almost every industry, from business to education. Emails can be categorized into two categories: ham and spam. Junk emails, also known as spam messages, are emails that have been designed to harm recipients by wasting their time, computing resources, and stealing their valuable information. It is estimated that spam emails are increasing at a rapid rate. One of the most important and prominent spam prevention techniques is filtering email. Naive Bayes, Decision Trees, Neural Networks, and Random Forests are among the methods used for this purpose by researchers. In this project, I examine the Logistic Regression machine learning model for spam filtering in email by categorizing messages into appropriate groups. This study also compares the techniques based on accuracy, precision, recall, etc. The accuracy level for this project was around 97%. Towards the end, these insights and future research directions, and challenges are outlined.

View PDFchevron_right
Using Data Mining Techniques and the Choice of Mode of Text Representation for Improving the Detection and Filtering of Spam
reda mohamed hamou

2015

This chapter studies a boosting algorithm based, first, on Bayesian filters that work by establishing a correlation between the presence of certain elements in a message and the fact that they appear in general unsolicited messages (spam) or in legitimate email (ham) to calculate the probability that the message is spam and, second, on an unsupervised learning algorithm: in this case the K-means. A probabilistic technique is used to weight the terms of the matrix term-category, and K-means are used to filter the two classes (spam and ham). To determine the sensitive parameters that improve the classifications, the authors study the content of the messages by using a representation of messages by the n-gram words and characters independent of languages to later decide what representation ought to get a good classification. The work was validated by several validation measures based on recall and precision.

View PDFchevron_right
A Comparative Study of Various Supervised Feature Selection Methods for Spam Classification
Shrawan Trivedi, Shubhamoy Dey

Classification of the spam from bunch of the email files is a challenging research area in text mining domain. However, machine learning based approaches are widely experimented in the literature with enormous success. For excellent learning of the classifiers, few numbers of informative features are important. This researh presents a comparative study between various supervised feature selection methods such as Document Frequency (DF), Chi-Squared (2 ), Information Gain (IG), Gain Ratio (GR), Relief F (RF), and One R (OR). Two corpuses (Enron and SpamAssassin) are selected for this study where enron is main corpus and spamAssassin is used for validation of the results. Bayesian Classifier is taken to classify the given corpuses with the help of features selected by above feature selection techniques. Results of this study shows that RF is the excellent feature selection technique amongst other in terms of classification accuracy and false positive rate whereas DF and 2  were not so effective methods. Bayesian classifier has proven its worth in this study in terms of good performance accuracy and low false positives.

View PDFchevron_right
Spam Mail Classifier
IJRASET Publication

International Journal for Research in Applied Science & Engineering Technology (IJRASET), 2022

Email is the worldwide use of communication application. It is because of the ease of use and faster than other communication application. However, its inability to detect whether the mail content is either spam or ham degrade its performance. Nowadays, lot of cases have been reported regarding stealing of personal information or phishing activities via email from the user. This project will discuss how machine learning help in spam detection. Machine learning is an artificial intelligence application that provides the ability to automatically learn and improve data without being explicitly programmed. Binary classifier will be used to classify the text into two different categories: spam and ham. The algorithm will predict the score more accurately. The objective of developing this model is to detect and score word faster and accurately.

View PDFchevron_right
Analysis of Random Forest and Naïve Bayes for Spam Mail using Feature Selection Catagorization
Rachana Mishra

International Journal of Computer Applications, 2013

Today, internet users are increases Spam mail is the major problem and big challenges for researcher to reduce it .Spam is commonly defined as unsolicited email messages and the goal of spam categorization is to distinguish between spam and legitimate email messages. This paper shows classification of spam mail and solving various problems is related to web space. Many machine learning algorithm are used to classified the spam and legitimate mail. This paper identify the best classification approach using bench mark dataset .The dataset consist of 9324 records and 500 attributes used for (training and testing) to build the model. This paper can play significant role to help eliminate unsolicited commercial e-mail, viruses, Trojans, and worms, as well as frauds perpetrated electronically and other undesired and troublesome e-mail. Three machines learning supervised algorithms namely naive bayes, Random Tree and Random Forest have applied on spam mail dataset using two feature selection algorithms.

View PDFchevron_right
Machine Learning Based Classification for Spam Detection
Onur Sevli

Sakarya Üniversitesi Fen Bilimleri Enstitüsü dergisi/Sakarya Üniversitesi fen bilimleri enstitüsü dergisi, 2023

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. 1. Introduction With the widespread use of the Internet, electronic communication has become more preferred. One of the most important tools of electronic communication is electronic messages, which we call e-mail. Today, individuals or organizations have one or more email accounts. Instant delivery of messages, no cost and ease of use increase the importance and prevalence of e-mail [1]. According to Statista Research Department data, the number of actively used e-mail accounts in 2020 is more than 4 billion. This number is estimated to increase to 4.6 billion in 2025. In 2020, 306 billion e-mails are sent and received every day, and this number is expected to exceed 376 billion in 2025 [2]. The use of e-mail is not only practical but also has various vulnerabilities. The e-mail account to be hijacked in various ways, for e-mails containing advertisements etc. to hijack your computer by installing a software on your computer when you click on the advertisement, and for the installed software to disrupt communication by sometimes filling the

View PDFchevron_right
Spam Detection Using Machine Learning
manju more

Computer Engineering and Intelligent Systems, 2020

Emails are essential in present century communication however spam emails have contributed negatively to the success of such communication. Studies have been conducted to classify messages in an effort to distinguish between ham and spam email by building an efficient and sensitive classification model with high accuracy and low false positive rate. Regular rule-based classifiers have been overwhelmed and less effective by the geometric growth in spam messages, hence the need to develop a more reliable and robust model. Classification methods employed includes SVM (support vector machine), Bayesian, Naïve Bayes, Bayesian with Adaboost, Naïve Bayes with Adaboost. However, for this project, the Bayesian was employed using Python programming language to develop a classification model.

View PDFchevron_right

References (22)

  1. S. A. A. Ghaleb et al., "Feature Selection by Multiobjective Optimization: Application to Spam Detection System by Neural Networks and Grasshopper Optimization Algorithm," in IEEE Access, vol. 10, pp. 98475-98489, 2022.
  2. Karim, S. Azam, B. Shanmugam, K. Kannoorpatti and M. Alazab, "A Comprehensive Survey for Intelligent Spam Email Detection," in IEEE Access, vol. 7, pp. 168261-168295, 2019.
  3. Karim, S. Azam, B. Shanmugam and K. Kannoorpatti, "Efficient Clustering of Emails Into Spam and Ham: The Foundational Study of a Comprehensive Unsupervised Framework," in IEEE Access, vol. 8, pp. 154759-154788, 2020.
  4. Karim, S. Azam, B. Shanmugam and K. Kannoorpatti, "An Unsupervised Approach for Content-Based Clustering of Emails Into Spam and Ham Through Multiangular Feature Formulation," in IEEE Access, vol. 9, pp. 135186-135209, 2021.
  5. G. Al-Rawashdeh, R. Mamat and N. Hafhizah Binti Abd Rahim, "Hybrid Water Cycle Optimization Algorithm With Simulated Annealing for Spam E-mail Detection," in IEEE Access, vol. 7, pp. 143721-143734, 2019.
  6. S. Gibson, B. Issac, L. Zhang and S. M. Jacob, "Detecting Spam Email With Machine Learning Optimized With Bio-Inspired Metaheuristic Algorithms," in IEEE Access, vol. 8, pp. 187914- 187932, 2020.
  7. S. Maroofi, M. Korczyński, A. Hölzel and A. Duda, "Adoption of Email Anti-Spoofing Schemes: A Large Scale Analysis," in IEEE Transactions on Network and Service Management, vol. 18, no. 3, pp. 3184-3196, Sept. 2021.
  8. S. A. A. Ghaleb, M. Mohamad, S. A. Fadzli and W. A. H. M. Ghanem, "Training Neural Networks by Enhance Grasshopper Optimization Algorithm for Spam Detection System," in IEEE Access, vol. 9, pp. 116768-116813, 2021.
  9. Z. -Y. Zhao and P. Zeng, "Efficient All-or-Nothing Public Key Encryption With Authenticated Equality Test," in IEEE Access, vol. 9, pp. 94099-94108, 2021.
  10. M. Hijji and G. Alam, "A Multivocal Literature Review on Growing Social Engineering Based Cyber-Attacks/Threats During the COVID-19 Pandemic: Challenges and Prospective Solutions," in IEEE Access, vol. 9, pp. 7152-7169, 2021.
  11. M. RAZA, N. D. Jayasinghe and M. M. A. Muslam, "A Comprehensive Review on Email Spam Classification using Machine Learning Algorithms," 2021 International Conference on Information Networking (ICOIN), Jeju Island, Korea (South), 2021, pp. 327-332.
  12. R. Wongwatkit, M. Raktham and T. Phawananthaphuti, "Intelligent Blacklist Security System for Protecting Spammer in Corporate Email Solution: A Case of Corporate Email Service Provider in Thailand," 2021 23rd International Conference on Advanced Communication Technology (ICACT), PyeongChang, Korea (South), 2021, pp. 387-391.
  13. C. Bansal and B. Sidhu, "Machine Learning based Hybrid Approach for Email Spam Detection," 2021 9th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO), Noida, India, 2021, pp. 1-4.
  14. A. Sumithra, A. Ashifa, S. Harini and N. Kumaresan, "Probability-based Naïve Bayes Algorithm for Email Spam Classification," 2022 International Conference on Computer Communication and Informatics (ICCCI), Coimbatore, India, 2022, pp. 1-5.
  15. S. Kaddoura, O. Alfandi and N. Dahmani, "A Spam Email Detection Mechanism for English Language Text Emails Using Deep Learning Approach," 2020 IEEE 29th International Conference on Enabling Technologies: Infrastructure for Collaborative Enterprises (WETICE), Bayonne, France, 2020, pp. 193-198.
  16. N. Gavrilovic and V. Ciric, "Design and Evaluation of Proof of Work Based Anti-Spam Solution," 2020 Zooming Innovation in Consumer Technologies Conference (ZINC), Novi Sad, Serbia, 2020, pp. 286-289.
  17. N. A. Farahisya and F. A. Bachtiar, "Spam Email Detection with Affect Intensities using Recurrent Neural Network Algorithm," 2022 2nd International Conference on Information Technology and Education (ICIT&E), Malang, Indonesia, 2022, pp. 206- 211.
  18. H. Dinendra, C. Rajapakse and P. P. G. Dinesh Asanka, "Personalized Classification of Non-Spam Emails Using Machine Learning Techniques," 2022 International Research Conference on Smart Computing and Systems Engineering (SCSE), Colombo, Sri Lanka, 2022, pp. 171-177.
  19. S. K. Ganiev and S. J. Khamidov, "Artificial Intelligence-Based Methods For Filtering Spam Messages In Email Services," 2021 International Conference on Information Science and Communications Technologies (ICISCT), Tashkent, Uzbekistan, 2021, pp. 1-4.
  20. A. B, D. P. M, K. M, M. N. Joseph and D. R, "Spam Detection System Using Supervised ML," 2021 International Conference on System, Computation, Automation and Networking (ICSCAN), 2021, pp. 1-5, doi: 10.1109/ICSCAN53069.2021.9526421. Authorized licensed use limited to: Charles Darwin University. Downloaded on May 27,2023 at 00:46:30 UTC from IEEE Xplore. Restrictions apply.
  21. Rundong Yang, Kangfeng Zheng, Bin Wu, Chunhua Wu, and Xiujuan Wang. Phishing website detection based on deep convolutional neural network and random forest ensemble learning. Sensors, 21(24):8281, 2021.
  22. Rayan, A. (2022). Analysis of e-Mail Spam Detection Using a Novel Machine Learning-Based Hybrid Bagging Technique. Computational Intelligence and Neuroscience, 2022.

Related papers

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

International Journal for Research in Applied Science and Engineering Technology (IJRASET), 2022

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.

View PDFchevron_right
Developing a Spam Email Detector
Saadya Fahad

2015

16 Abstract— Email is obviously important for many types of group communication that have become most widely used by millions of people, individuals and organizations. At the same time it has become a prone to threats. The most popular such threats is what is called a spam, also known as unsolicited bulk email or junk email. To detect spams, this work proposes a spam detection approach using Naive Bayesian (NB) classifier, where this classifier identifies email messages as being spam or legitimate, based on the content (i.e. body) of these messages. Each email is represented as a bag of its body’s words (features). To catch up with the spammers latest techniques, a robust, yet up-to-dated dataset CSDMC2010 spam corpus (last updated 2014) a set of raw email messages, was considered. To best perform, NB’s environment was integrated with a list of 149 features proposed to include those commonly used by most spam emails. CSDMC2010 dataset was used to train and test NB classifier. Certai...

View PDFchevron_right
Analysis of Spam Messages Using Various Machine Learning Classifier
Joyece Jane

Background: As people using social media increases the data generation also increases and the data generated may be safe or unsafe. If we see some applications like Twitter and mail. We get a lot of emails or twits that include all dangerous and useful things. Here to be safe from the threats and dangers we need a filter that separates useful messages from spam and helps us not to drown in a trap. And one of the approaches to do this is explained in this paper. In this paper, the algorithm followed is the Naïve Bayes classifier. This also provides the comparison between using Naïve Bayes, KNN, and Logistic Regression to solve the same problem that is spam filtering and term frequency-inverse document frequency (TFIDF).

View PDFchevron_right
Effect of Header-based Features on Accuracy of Classifiers for Spam Email Classification
Haridasa Acharya

International Journal of Advanced Computer Science and Applications, 2020

Emails are an integral part of communication in today's world. But Spam emails are a hindrance, leading to reduction in efficiency, security threats and wastage of bandwidth. Hence, they need to be filtered at the first filtering station, so that employees are spared the drudgery of handling them. Most of the earlier approaches are mainly focused on building content-based filters using body of an email message. Use of selected header features to filter spam, is a better strategy, which was initiated by few researchers. In this context, our research intends to find out minimum number of features required to classify spam and ham emails. A set of experiments was conducted with three datasets and five Feature Selection techniques namely Chi-square, Correlation, Relief Feature Selection, Information Gain, and Wrapper. Five-classification algorithms-Naïve Bayes, Decision Tree, NBTree, Random Forest and Support Vector Machine were used. In most of the approaches, a trade-off exists between improper filtering and number of features. Hence arriving at an optimum set of features is a challenge. Our results show that in order to achieve the objective of satisfactory filtering, minimum 5 and maximum 14 features are required.

View PDFchevron_right
Highly Accurate Spam Detection with the Help of Feature Selection and Data Transformation
Fatema Nusrat

The International Arab Journal of Information Technology

The amount of spam is increasing rapidly while the popularity of emails is increasing. This situation has led to the need to filter spam emails. To date, many knowledge-based, learning-based, and clustering-based methods have been developed for filtering spam emails. In this study, machine-learning-based spam detection was targeted, and C4.5, ID3, RndTree, C-Support Vector Classification (C-SVC), and Naïve Bayes algorithms were used for email spam detection. In addition, feature selection and data transformation methods were used to increase spam detection success. Experiments were performed on the UC Irvine Machine Learning Repository (UCI) spambase dataset, and the results were compared for accuracy, Receiver Operating Characteristic (ROC) analysis, and classification speed. According to the accuracy comparison, the C-SVC algorithm gave the highest accuracy with 93.13%, followed by the RndTree algorithm. According to the ROC analysis, the RndTree algorithm gave the best Area Under...

View PDFchevron_right

Related topics

  • Machine Learning
    • 580 California St., Suite 400
    San Francisco, CA, 94104
    © 2025 Academia. All rights reserved