Email Classification Research Articles

Toward Accurate Feature Selection Based on BSS-GRF

As of late, Feature extraction in email classification assumes a vital part. Many Feature extraction algorithms need more effort in term of accuracy. In order to improve the classifier accuracy and for faster classification, the hybrid algorithm is proposed. This hybrid algorithm combines the Genetics Rough set with blind source separation approach (BSS-GRF). The main aim of proposing this hybrid algorithm is to improve the classifier accuracy for classifying incoming e-mails.

Classification of Phishing Email Using Random Forest Machine Learning Technique

Phishing is one of the major challenges faced by the world of e-commerce today. Thanks to phishing attacks, billions of dollars have been lost by many companies and individuals. In 2012, an online report put the loss due to phishing attack at about $1.5 billion. This global impact of phishing attacks will continue to be on the increase and thus requires more efficient phishing detection techniques to curb the menace. This paper investigates and reports the use of random forest machine learning algorithm in classification of phishing attacks, with the major objective of developing an improved phishing email classifier with better prediction accuracy and fewer numbers of features. From a dataset consisting of 2000 phishing and ham emails, a set of prominent phishing email features (identified from the literature) were extracted and used by the machine learning algorithm with a resulting classification accuracy of 99.7% and low false negative (FN) and false positive (FP) rates.

Authorship Attribution in Tamil Language Email for Forensic Analysis

This paper presents Authorship attribution (AA) to Tamil language email. This work presents generation of representative signatures of Tamil emails using lexical and syntactic based methods. The signature of each email has large dimensions. In order to make it suitable for subsequent processing, conversion of large dimension of the signature into 2-dimensional pattern using Fisher’s linear discriminant function (FLD) method is given. The 2-dimensional patterns of the signatures are used as training data for the radial basis function (RBF) network and echo state neural network (ESNN). The improved classification of Tamil email is shown by transformation of patterns using FLD followed by training using RBF, as well as, training ESNN. The paper presents a new technique for building signature database and for optimal AA in Tamil email forensics.

Spam Mail Detection through Data Mining – A Comparative Performance Analysis

As web is expanding day by day and people generally rely on web for communication so e-mails are the fastest way to send information from one place to another. Now a day's all the transactions all the communication whether general or of business taking place through e-mails. E-mail is an effective tool for communication as it saves a lot of time and cost. But e- mails are also affected by attacks which include Spam Mails. Spam is the use of electronic messaging systems to send bulk data. Spam is flooding the Internet with many copies of the same message, in an attempt to force the message on people who would not otherwise choose to receive it. In this study, we analyze various data mining approach to spam dataset in order to find out the best classifier for email classification. In this paper we analyze the performance of various classifiers with feature selection algorithm and without feature selection algorithm. Initially we experiment with the entire dataset without selecting the features and apply classifiers one by one and check the results. Then we apply Best-First feature selection algorithm in order to select the desired features and then apply various classifiers for classification. In this study it has been found that results are improved in terms of accuracy when we embed feature selection process in the experiment. Finally we found Random Tree as best classifier for spam mail classification with accuracy = 99.72%. Still none of the algorithm achieves 100% accuracy in classifying spam emails but Random Tree is very nearby to that.

E-mail spam filtering by a new hybrid feature selection method using IG and CNB wrapper

The growing volume of spam emails has resulted in the necessity for more accurate and efficient email classification system. The purpose of this research is presenting an machine learning approach for enhancing the accuracy of automatic spam detecting and filtering and separating them from legitimate messages. In this regard, for reducing the error rate and increasing the efficiency, the hybrid architecture on feature selection has been used. Features used in these systems, are the body of text messages. Proposed system of this research has used the combination of two filtering models, Filter and Wrapper, with Information Gain (IG) filter and Complement Naïve Bayes (CNB) wrapper as feature selectors. In addition, Multinomial Naïve Bayes (MNB) classifier, Discriminative Multinomial Naïve Bayes (DMNB) classifier, Support Vector Machine (SVM) classifier and Random Forest classifier are used for classification. Finally, the output results of this classifiers and feature selection methods are examined and the best design is selected and it is compared with another similar works by considering different parameters. The optimal accuracy of the proposed system is evaluated equal to 99%.

End-user feature labeling: Supervised and semi-supervised approaches based on locally-weighted logistic regression

When intelligent interfaces, such as intelligent desktop assistants, email classifiers, and recommender systems, customize themselves to a particular end user, such customizations can decrease productivity and increase frustration due to inaccurate predictions—especially in early stages when training data is limited. The end user can improve the learning algorithm by tediously labeling a substantial amount of additional training data, but this takes time and is too ad hoc to target a particular area of inaccuracy. To solve this problem, we propose new supervised and semi-supervised learning algorithms based on locally-weighted logistic regression for feature labeling by end users, enabling them to point out which features are important for a class, rather than provide new training instances.We first evaluate our algorithms against other feature labeling algorithms under idealized conditions using feature labels generated by an oracle. In addition, another of our contributions is an evaluation of feature labeling algorithms under real-world conditions using feature labels harvested from actual end users in our user study. Our user study is the first statistical user study for feature labeling involving a large number of end users (43 participants), all of whom have no background in machine learning.Our supervised and semi-supervised algorithms were among the best performers when compared to other feature labeling algorithms in the idealized setting and they are also robust to poor quality feature labels provided by ordinary end users in our study. We also perform an analysis to investigate the relative gains of incorporating the different sources of knowledge available in the labeled training set, the feature labels and the unlabeled data. Together, our results strongly suggest that feature labeling by end users is both viable and effective for allowing end users to improve the learning algorithm behind their customized applications.

A Novel Technique of Email Classification for Spam Detection

Email spam is one of the unsolved problems of the today's Internet, annoying individual users and bringing financial damage to companies. Among the approaches developed to stop spam emails, filtering is a popular and important one. Common uses for email filters include organizing incoming email and computer viruses and removal of spam. As spammers periodically find new ways to bypass spam filters and distribute spam messages, researchers need to stay on the forefront of this problem to help reduce the amount of spam messages. Currently spam emails occupy more than 70% of all email traffic. The negative effect spam has on companies is greatly related to financial aspects and the productivity of employees in the workplace. In this paper, we propose the new approach to classify spam emails using support vector machine. It found that the SVM outperformed than other classifiers.

Email mining: tasks, common techniques, and tools

Email is one of the most popular forms of communication nowadays, mainly due to its efficiency, low cost, and compatibility of diversified types of information. In order to facilitate better usage of emails and explore business potentials in emailing, various data mining techniques have been applied on email data. In this paper, we present a brief survey of the major research efforts on email mining. To emphasize the differences between email mining and general text mining, we organize our survey on five major email mining tasks, namely spam detection, email categorization, contact analysis, email network property analysis and email visualization. Those tasks are inherently incorporated into various usages of emails. We systematically review the commonly used techniques and also discuss the related software tools available.

Tuned Artificial Neural Network Model for E-mail Data Classification with Feature Selection

With the rapid development of Internet, e-mail has become effective means of communication to share information. Through e-mail, we can send text messages, images, audio and video clips across the world within a fraction of time. In recent years, e-mail users are facing problem due to spam e-mails. Spam e-mails are unsolicited commercial/bulk e-mails sent by spammers. There are many serious problems associated with spam e-mails, e.g. it may contain hyperlink which may lead to a bogus website which might ask you for your personal information like username, password, bank account number etc.. Spam e-mail is not only wastage of storage space but also wastage of time. In order to tackle problems faced by users due to spam e-mail, it is necessary to classify them with the help of intelligent and robust classifier. These classifiers should have the capability to classify spam e-mail against non-spam e-mail. The spam e-mail classifier performance can be greatly enhanced with the use of artificial neural network classification algorithm. An Artificial Neural Network (ANN) is a powerful tool used for classification of data , it has capability of learning huge amount of data with high dimensionality in better way, there are various parameters of ANN to be set to tune for the better performance of neural network model, these are learning rate, architecture of ANN and momentum, these all parameters play a very important role in improving the accuracy of ANN model. In this paper Error Back Propagation Network (EBPN) techniques based on ANN are explored with different value of learning rate from 0.2 to 0.9. An EBPN model is derived from e-mail data set obtained from UCI repository site with three different partitions. Due to high dimensionality of data set, we have applied feature selection technique for the best model. This model is tested with various combinations of feature and it is concluded that model is producing highest accuracy of 98.49% on testing samples with 52 features. The derived model is also measured with precision, recall and F-measure and achieved 98.34%, 99.07% and 98.70% respectively.

A comparative study on feature selection and adaptive strategies for email foldering using the ABC-DynF framework

Email foldering is a challenging problem mainly due to its high dimensionality and dynamic nature. This work presents ABC-DynF, an adaptive learning framework with dynamic feature space that we use to compare several incremental and adaptive strategies to cope with these two difficulties. Several studies have been carried out using datasets from the ENRON email corpus and different configuration settings of the framework. The main aim is to study how feature ranking methods, concept drift monitoring, adaptive strategies and the implementation of a dynamic feature space can affect the performance of Bayesian email classification systems.

Enhancing Phishing E-Mail Classifiers: A Lexical URL Analysis Approach

This is a study that focuses on enhancing the mitigation of bulk phishing email messages (i.e. email messages with generic socially engineered content that target a broad range of recipients). This study is based on a phishing website detection technique that we have proposed previously. The previously proposed technique was able to achieve 97% of classification accuracy of phishing websites by lexically analyzing their URLs. The centre claim of this study is that the classification accuracy of anti-phishing email filters enhance when they incorporate the proposed lexical URL analysis technique. To evaluate the claims, a highly accurate anti-phishing email classifier is constructed and tested against publicly available phishing and legitimate email data sets.

Phishing Dynamic Evolving Neural Fuzzy Framework for Online Detection “Zero-day” Phishing Email

Phishing is a kind of attack in which criminals use spoofed emails and fraudulent web sites to trick financial organization and customers. Criminals try to lure online users by convincing them to reveal the username, passwords, credit card number and updating account information or fill billing information. One of the main problems of phishing email detection is the unknown “zero-day” phishing attack, (we define zero-day attacks as attacks that phisher mount using hosts that do not appear in blacklists and not trained on the old data sample and it is a noise data), which increases the level of difficulty to detect phishing email. Nowadays, phishers are creating different representation techniques to create unknown “zero-day” phishing email to breach the defenses of those detectors. Our proposed is a novel framework called phishing dynamic evolving neural fuzzy framework (PDENF), which adapts the evolving connectionist system (ECoS) based on a hybrid (supervised/unsupervised) learning approach. PDENF adaptive online is enhanced by offline learning to detect dynamically the phishing email included unknown zero-day phishing e-mails before it get to user account. PDENF is suggested to work for high-speed “life-long” learning with low memory footprint and minimizes the complexity of the rule base and configuration with few number of rules creation for email classification. We expect to achieves high performance, including high level of true positive, true negative, sensitivity, precision, F-measure and overall accuracy compared with other approaches.

The importance of context in the automatic classification of email as records of business value: A pilot study

AbstractDespite the growth of alternative means of electronic messaging, email continues to be a crucial business communication tool. In many organizations, it is common to find a large proportion of business‐critical data in email form. However, as the volume of business‐critical email continues to grow, conventional approaches for managing paper‐based records of business value are increasingly unsuited to the digital realities. Yet effective approaches to automate the management of email are essential to ensure organizational efficiency, accountability and regulatory compliance. Our research aims to develop a model that characterizes information managers' context‐specific email triage strategies for identifying emails of business value so that it can be applied to automatically classify email records. We conducted a pilot study of two information managers' email triage practices in two different contexts and develop machine‐learning models of lexical and non‐lexical features that are involved in the appraisal of business value. An experiment with a machine learning algorithm trained on about two hundred emails in two different business contexts indicates that the automation of email triage for identifying records of business value is highly context dependent and that automated classifiers must be trained to recognize “business value” specifically for that context.

Three-Phase Tournament-Based Method for Better Email Classification

Three-Phase Tournament-Based Method for Better Email Classification

Rough sets for spam filtering: Selecting appropriate decision rules for boundary e-mail classification

Nowadays, spam represents an extensive subset of the information delivered through Internet involving all unsolicited and disturbing communications received while using different services including e-mail, weblogs and forums. In this context, this paper reviews and brings together previous approaches and novel alternatives for applying rough set (RS) theory to the spam filtering domain by defining three different rule execution schemes: MFD (most frequent decision), LNO (largest number of objects) and LTS (largest total strength). With the goal of correctly assessing the suitability of the proposed algorithms, we specifically address and analyse significant questions for appropriate model validation like corpus selection, preprocessing and representational issues, as well as different specific benchmarking measures. From the experiments carried out using several execution schemes for selecting appropriate decision rules generated by rough sets, we conclude that the proposed algorithms can outperform other well-known anti-spam filtering techniques such as support vector machines (SVM), Adaboost and different types of Bayes classifiers.

Training SVM email classifiers using very large imbalanced dataset

The Internet has been flooded with spam emails, and during the last decade there has been an increasing demand for reliable anti-spam email filters. The problem of filtering emails can be considered as a classification problem in the field of supervised learning. Theoretically, many mature technologies, for example, support vector machines (SVM), can be used to solve this problem. However, in real enterprise applications, the training data are typically collected via honeypots and thus are always of huge amounts and highly biased towards spam emails. This challenges both efficiency and effectiveness of conventional technologies. In this article, we propose an undersampling method to compress and balance the training set used for the conventional SVM classifier with minimal information loss. The key observation is that we can make a trade-off between training set size and information loss by carefully defining a similarity measure between data samples. Our experiments show that the SVM classifier provides a better performance by applying our compressing and balancing approach.

Email pragmatics and automatic classification: A study in the organizational context

This paper presents a two-phased research project aiming to improve email triage for public administration managers. The first phase developed a typology of email classification patterns through a qualitative study involving 34 participants. Inspired by the fields of pragmatics and speech act theory, this typology comprising four top level categories and 13 subcategories represents the typical email triage behaviors of managers in an organizational context. The second study phase was conducted on a corpus of 1,703 messages using email samples of two managers. Using the k-NN (k-nearest neighbor) algorithm, statistical treatments automatically classified the email according to lexical and nonlexical features representative of managers' triage patterns. The automatic classification of email according to the lexicon of the messages was found to be substantially more efficient when k = 2 and n = 2,000. For four categories, the average recall rate was 94.32%, the average precision rate was 94.50%, and the accuracy rate was 94.54%. For 13 categories, the average recall rate was 91.09%, the average precision rate was 84.18%, and the accuracy rate was 88.70%. It appears that a message's nonlexical features are also deeply influenced by email pragmatics. Features related to the recipient and the sender were the most relevant for characterizing email.

Social Network Based Reputation Computation and Document Classification

We develop two social network based algorithms that automatically com- pute author reputation from a collection of textual documents. We first extract keyword reference behaviors of the authors to construct a social network, which represents re- lationships among the authors in terms of information reference behavior .W ith this network, we apply the two algorithms: the first computes each author's reputation value considering only direct reference and the second utilizes indirect reference recursively. We compare the reputation values computed by the two algorithms and reputation ratings given by a human domain expert. We further evaluate the algorithms in email categorization tasks by comparing them with machine learning techniques. Finally, we analyse the social network through a community detection algorithm and other analy- sis techniques. We observed several interesting phenomena including the network being scale-free and having a negative assortativity.

Cognitive Category Learning

Categories play a fundamental role in our daily lives and are the basis for decision making in most professions (e.g., intelligence analysis, healthcare, and engineering). Categories are a set of objects or events that have similar features and are grouped together because of their similarity. Many categories are acquired as a child (e.g., tools, fruit) but we continue to learn and apply new categories throughout life. Categories range from concrete (e.g., a set of physical objects like houses, dogs…) to abstract (e.g., political ideologies, movie genres…) and narrow (e.g., rifles) to broad (e.g., weapons). People in a variety of knowledge intensive fields (e.g., analysts, commanders, medical doctors) recognize categories in streams of data and make a decision about how to act (or pass the information to a decision maker). Such a categorization system is important now, especially because of the large amount of information that people need to sift through to do their jobs on a regular basis. When decision support systems (DSSs) are applied to support human decision making by automatically recognizing categories, these systems are often not practical in complex dynamic real world environments. Rule based DSSs are limited because it is difficult to develop and maintain large complex rule sets. Current machine learning based DSSs are sometimes limited because they require data that encompasses all of the possible variations as examples to learn from and necessitate significant effort to develop and maintain this training data. This paper describes a cognitive category learning system that uses machine learning and natural language processing (NLP) techniques to categorize unstructured documents or semi-structured objects, such as emails, which we used in this experiment. Our system uses several methods to do this categorization of emails and then arbitrates the best solution based on the individual classifier results. This result provides a more confident answer with less chance of false positive and false negative outcomes. Our system also generates a metadata topic summary for each document or email.

Social feature-based enterprise email classification without examining email contents

Without imposing restrictions, many enterprises find nonwork-related contents consuming network resources. Business communication over emails thus incurs undesired delays and inflicts damages to businesses, explaining why many enterprises are concerned with the competition to use email services. Obviously, enterprises should prioritize business emails over personal ones in their email service. Therefore, previous works present content-based classification methods to categorize enterprise emails into business or personal correspondence. Accuracy of these methods is largely determined by their ability to survey as much information as possible. However, in addition to decreasing the performance of these methods, monitoring the details of email contents may violate privacy rights that are under legal protection, requiring a careful balance of accurately classifying enterprise emails and protecting privacy rights. The proposed email classification method is thus based on social features rather than a survey of emails contents. Social-based metrics are also designed to characterize emails as social features; the obtained features are treated as an input of machine learning-based classifiers for email classification. Experimental results demonstrate the high accuracy of the proposed method in classifying emails. In contrast with other content-based methods that examine email contents, the emphasis on social features in the proposed method is a promising alternative for solving similar email classification problems.