Email Spam Classification Research Articles

Email Spam Classification Using Support Vector Machine Algorithm

Email Spam Classification Using Support Vector Machine Algorithm

Ham and Spam E-Mails Classification Using Machine Learning Techniques

ABSTRACTSpam e-mail has become a very serious problem. Sending inappropriate messages to a large number of recipients indiscriminately has resulted in anger by users but large profits for spammers. This article looks at classifying spam e-mails from inboxes. Ten alternative classifiers are applied on one benchmark dataset to evaluate which classifier gives better result. A 10-fold cross validation is used to provide the accuracy. Results of the classification algorithms are compared with the spambase UCI dataset. The experimental results approve that the spam mails can be classified correctly, with accuracy reaching up to 95.45% for the Random Forest technique, compared to other classifiers used.

E-Mail Spam Classification Using Machine Learning in Distributed Environment

E-Mail Spam Classification Using Machine Learning in Distributed Environment

Random Forests Spam Email Classification System

Email is a rapid communication tool and cost-effective for the users. Conversely, the number of email users driven to the intense growth of spam mails in the past few eras. This spam mails issue is one of the substantial risks with the internet. The growing amount of spam mails brings the importance of trustworthy anti-spam filters. Usually the spammers send the undesired and unsolicited emails to various recipients and these spam mails are mostly identical in its characteristics. Hence it is essential to frame a defense system that effectively finds the spam mails and provide an alternate process for stand-alone filter. Thus, in this paper a novel framework for classifying the email into spam and ham mails using the attribute based random forests classification is proposed. The process begins Bayesian spamminess probability calculation for each token, TF-IDF weighting scheme calculates the weight for each token and the mail, score calculation is performed based on the genetic fitness and finally the classification process is done using random forests classifier to classify the emails into spam and ham emails The results are compared with existing spam classification methods in terms of classification accuracy, weighted accuracy and F1 measure. The results show that the proposed system shows the promising results when compared with other existing algorithms.

Visual and textual features based email spam classification using S-Cuckoo search and hybrid kernel support vector machine

Spam mail classification has been playing a vital role in recent days due to the uncontrollable growth happening in the electronic media. Literature presents several algorithms for email spam classification based on classification methods. In this paper, we propose a spam classification framework using S-Cuckoo and hybrid kernel based support vector machine (HKSVM). At first, the features are extracted from the e-mails based on the text as well as the image. For the textual features, TF-term frequency is used. For the image dependent features, correrlogram and wavelet moment are taken. The hybrid features have then high dimension so the optimum features are identified with the help of hybrid algorithm, called S-Cuckoo search. Then, the classification is done using proposed classifier HKSVM model which is designed based on the hybrid kernel by blending three different kernel functions and then it is used in the SVM classifier. The additional features provided based on image and the modification of SVM classifier provides significant improvement as compared with existing algorithms. The spam classification performance is measured by db1 (combining bare-ling spam and Spam Archive corpus) and db2 (combining lemm-ling spam and Spam Archive corpus). Experimental results show that the proposed spam classification framework has outperformed by having better accuracy of 97.235% when compared with existing approach which is able to achieve only 94.117%.

An Ensembled Classifier for Email Spam Classification in Hadoop Environment

An Ensembled Classifier for Email Spam Classification in Hadoop Environment

E-mail spam classification using S-cuckoo search and support vector machine

Today's internet world email spam becoming a major problem to the internet users. The main intention of this paper is to design a technique for email spam classification using modified cuckoo search called as stepsize-cuckoo search (SCS) and support vector machine. Optimised feature set is identified with the use of SCS algorithm. Once the best feature set is identified through SCS, the spam classification is done using the support vector machine. For the effectiveness of classification, we have used three different kernels such as linear, polynomial and quadratic. To evaluate the proposed email spam classification, we have used evaluation metrics such as sensitivity, specificity and accuracy. From the results, it clearly shows that our method has shown higher accuracy than the original cuckoo with SVM for spam base datasets.

A Novel Weak Estimator For Dynamic Systems

In this paper, we propose a novel approach for classifying incoming continuous data under a non-stationary environment. A class of estimators termed stochastic learning weak estimators has been generalized to include continuous time sampling and countable state categories. The method is founded on non-stationary Markov chain techniques and is useful in diverse applications, such as consumer behavior analysis, e-mail spam classification, or understanding drug effectiveness. In terms of tracking the true state probabilities, these weak estimators consistently outperform traditional competitors such as maximum likelihood estimates. Only one user defined parameter is necessary and the method is free of subjective “moving window” type algorithms. We have conducted extensive simulations and real data analyses for classification purposes.

E-mail spam classification using S-cuckoo search and support vector machine

E-mail spam classification using S-cuckoo search and support vector machine

Email Spam Classification in a Distributed Environment

In today's computer world, Internet plays a main role. Internet mail system is a store and forward mechanism used for the purpose of exchanging documents across computer network through Internet. Unsolicited bulk e-mail known as spam mail is considered to be a major threat. The objective of the proposed system is to develop a spam classification system in a distributed environment using Naive bayes and SVM classifier. Performance analysis of the above two classifiers is analyzed under the distributed environment. Implementation of distributed environment using Hadoop's Map Reduce framework serves to be more efficient which gives good results for the above system that has been proposed in this paper.

Email Classification Research Trends: Review and Open Issues

Personal and business users prefer to use e-mail as one of the crucial sources of communication. The usage and importance of e-mails continuously grow despite the prevalence of alternative means, such as electronic messages, mobile applications, and social networks. As the volume of business-critical e-mails continues to grow, the need to automate the management of e-mails increases for several reasons, such as spam e-mail classification, phishing e-mail classification, and multi-folder categorization, among others. This paper comprehensively reviews articles on e-mail classification published in 2006–2016 by exploiting the methodological decision analysis in five aspects, namely, e-mail classification application areas, data sets used in each application area, feature space utilized in each application area, e-mail classification techniques, and the use of performance measures. A total of 98 articles (56 articles from Web of Science core collection databases and 42 articles from Scopus database) are selected. To achieve the objective of the study, a comprehensive review and analysis is conducted to explore the various areas where e-mail classification was applied. Moreover, various public data sets, features sets, classification techniques, and performance measures are examined and used in each identified application area. This review identifies five application areas of e-mail classification. The most widely used data sets, features sets, classification techniques, and performance measures are found in the identified application areas. The extensive use of these popular data sets, features sets, classification techniques, and performance measures is discussed and justified. The research directions, research challenges, and open issues in the field of e-mail classification are also presented for future researchers.

Weighted-based multiple classifier and F-GSO algorithm for email spam classification

In this paper, an efficient spam classification technique is proposed using weighted-based multiple classifier and F-GSO algorithm. At first, input data is given to the feature selection to select the suitable feature for spam classification. Here, firefly and GSO algorithm is effectively hybridised to select the suitable features. Once the best feature is identified through hybrid algorithm, the spam classification is done using the weighted-based multiple classifiers. Here, three categories of classifiers like, rule classifier, lazy classifier and learning classifiers is combined using weight rule. These three classifiers have their own advantages and disadvantages so the hybridisation of classifiers leads to provide overall improvements by rectifying their disadvantages by other algorithms and retaining their advantages. Accordingly, decision tree (rule), lazy classifier (naïve Bayes) and neural network classifier (learning) are combined using voting-based weighted rule. Our experiment result shows the proposed systems have outperformed by having better accuracy value of 98.83%.

Email Spam Classification using Hybridized Technique with Feature Selection

Email Spam Classification using Hybridized Technique with Feature Selection

Novel Email Spam Classification using Integrated Particle Swarm Optimization and J48

E-mails have become an integral part of both private and professional lives and can also be studied as formal papers in communication between users. Several activities such as spam detection and classification, subject classification, etc. can be done by email’s data mining and analysis. Review has shown that the use of unsupervised filtering to filter the input data set is ignored by the most of the existing researchers. The use of hybridization of data mining techniques is ignored in order to improve the accuracy rate further for detection of fraudulent emails. Most of the existing techniques are limited to some significant features of emails therefore utilising more features may provide more significant results. The overall objective of this work is to propose an integrated particle swarm optimization based J48 algorithm to enhance the accuracy rate further.

Data Mining for Spam Email Classification

This paper proposes a classification model for spam email using data mining. The use of personal or business email has increased along with the growth of internet population and computerization. This change allows one to live in convenience or to improve the efficiency of business. On the other hand, spam emails cause many problems in our life. Spam email is defined as the email which is sent to anyone who does not want to receive the email that brings to annoyance or computer virus or interruption of business process. Although a lot of studies have been proposed to protect the spam email, we are still in need of an efficient classification method. Data mining is a prominent way to classify spam emails. However, if data do not have sufficient information, traditional data mining method may not apply for the problem. Therefore, we suggest PU learning algorithm to classify the problem with insufficient data which have only positive class and unlabeled data. Support vector machine (SVM) has been used as the basic data mining method. Experimental results show the viability of the proposed classification model.

Classifying Spam Emails Using Artificial Intelligent Techniques

Spam emails have become an increasing difficulty for the entire web-users.These unsolicited messages waste the resources of network unnecessarily. Customarily, machine learning techniques are adopted for filtering email spam. This article examines the capabilities of the extreme learning machine (ELM) and support vector machine (SVM) for the classification of spam emails with the class level (d). The ELM method is an efficient model based on single layer feed-forward neural network, which can choose weights from hidden layers,randomly. Support vector machine is a strong statistical learning theory used frequently for classification. The performance of ELM has been compared with SVM. The comparative study examines accuracy, precision, recall, false positive, true positive.Moreover, a sensitivity analysis has been performed by ELM and SVM for spam email classification.

Data Mining for Spam Email Classification

Data Mining for Spam Email Classification

Supervised classification of spam emails with natural language stylometry

Email spam is one of the biggest threats to today's Internet. To deal with this threat, there are long-established measures like supervised anti-spam filters. In this paper, we report the developme...

IMPROVED SPAMBASE DATASET PREDICTION USING SVM RBF KERNEL WITH ADAPTIVE BOOST

Spam is no more garbage but risk as it includes virus attachments and spyware agents which make the recipients’ system ruined, therefore, there is an emerging need for spam detection. Many spam detection techniques based on machine learning algorithms have been proposed. As the amount of spam has been increased tremendously using bulk mailing tools, spam detection techniques should deal with it. In this paper we have proposed Hybrid classifier Adaptive boost with support vector machine RBF kernel on Spambase dataset. We have also extracted the features first by Principal component analysis. General Terms: Email Spam classification.

Improving the Accuracy of Diabetes Classification Using Modified K-Means

This research work has presented different data mining algorithms for efficiently classify the diabetes data. Data mining algorithms have becoming popular in real time applications like e-mail spam classification, intrusion detection system, diabetes mining, etc. The overall goal of this research work is to evaluate the performance of the K-means based data mining algorithms by considering the diabetes data set. As the accuracy rate of traditional K-means algorithm is 76% that means comes up with 24% error rate. So it is required to improve this accuracy rate. To achieve this objective we have modified the pre-processing of data set by using the well-known tool MATLAB. MATLAB has used WEKA machine learning tool as an API (Application programming interface). It has been found that that the Modified K-Means outperforms over the existing machine learning algorithms.