Email Classification

When email became more widely used, in the late 1990s, I heard the term “snail mail” to refer to mail by post. Whereas anyone online could almost instantly send a little letter to someone else online, or a big one, if one uses traditional post it takes at least 12 hours to arrive, if not longer. But is email always faster than post, when it is logically and naturally possible than it can be? (There is no logical inconsistency and there is no violation of a law of nature: a miracle is not required.)



Imagine that one finds that for some items, the post takes a mysteriously long time to arrive. I found this when ordering old copies of psychoanalytic journals, one of which wrote of “conceptual scheme bs.” One concludes that there is some reason for this delay, and even constructs an economic model in which it is rational for post to arrive slowly in this case: slower than logically and naturally possible. But will not email now or soon be affected by the same model? After all, who provides us with email? Organizations, which must take into account economic good sense as well. Email might work instantly for you, but if you seek the kind of information I sought when ordering these journals, then you may well face the same problem.



But why  does email arrive slowly in these case? I am not always sure. Even if one does not construct an economic model at all for snail mail, if one assumes that there is economic reason with some of the delays, then one has reason to assume that these will occur with email as well.

Electronic Mail (E-mail) has established a significant place in information user’s life. Mails are used as a major and important mode of information sharing because emails are faster and effective way of communication. Email plays its important role of communication in both personal and professional aspects of one’s life. The rapid increase in the number of account holders from last few decades and the increase in the volume of mails have generated various serious issues too. The content base mail classification can be classified into four ways namely Private, Public, Newsletter, and Anonymous. Every user has the right to choose their keyword (a semi-private password). Those contacts who know the user’s keyword will be classified as private contacts and those users who are unknown them classified anonymous contacts. A contact can be classified as public or private, upon verification of an anonymous contact. Any newsletter or group mails are classified into newsletter contacts. It is...

In the real world, many online shopping websites or service provider have single email-id where customers can send their query, concern etc. At the back-end service provider receive million of emails every week, how they can identify which email is belonged of a particular department? This paper presents an artificial neural network (ANN) model that is used to solve this problem and experiments are carried out on user personal Gmail emails datasets. This problem can be generalised as typical Text Classification or Categorization [8].

Trees have been a crucial component in humans' lives for hundreds of years, providing food, shelter, and medicine. Some trees have a lot of medicinal properties that cure many diseases. In the old days, Ayurvedic methods were popular for various treatments, but nowadays, the demand for foreign medicine is increasing gradually, which also has side effects. This paper addresses this issue by deciding on the medical conditions corresponding to a symptom and predicting an herb leaf that can be treated it using some modern machine learning techniques. We have used three machine learning methods to accomplish this goal: Multinomial Naive Bayes, Gradient Boosting and Random Forest. These techniques were then used to assess the symptoms and decide the name of the disease and which leaf is appropriate for medicine. The highest accuracy (92%) was produced by the Multinomial Naive Bayes algorithm, thereby showing its capability to predict the right medicinal leaf based on given symptoms. The results show that machine learning algorithms, especially Multinomial Naive Bayes, can identify diseases and recommend suitable medicinal leaves. This approach holds promise for integrating traditional Ayurvedic knowledge with modern technology to offer alternative treatments with potentially fewer side effects.

Modern email clients support predicate-based folder assignment rules that can automatically organize emails. Unfortunately, users still need to write these rules manually. Prior machine learning approaches have framed automatically assigning email to folders as a classification task and do not produce symbolic rules. Prior inductive logic programming (ILP) approaches, which generate symbolic rules, fail to learn efficiently in the online environment needed for email management. To close this gap, we present EmFore, an online system that learns symbolic rules for email classification from observations. Our key insights to do this successfully are: (1) learning rules over a folder abstraction that supports quickly determining candidate predicates to add or replace terms in a rule, (2) ensuring that rules remain consistent with historical assignments, (3) ranking rule updates based on existing predicate and folder name similarity, and (4) building a rule suppression model to avoid surfacing low-confidence folder predictions while keeping the rule for future use. We evaluate on two popular public email corpora and compare to 13 baselines, including state-of-the-art folder assignment systems, incremental machine learning, ILP and transformer-based approaches. We find that EmFore performs significantly better, updates four orders of magnitude faster, and is more robust than existing methods and baselines. CCS CONCEPTS • Computing methodologies → Online learning settings; Rule learning; • Information systems → Email.

Personal and business users prefer to use email 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. Finding out the important mails of the all the emails received on the same day is becoming difficult for many users, as the volume of critical emails continues to grow, the need to automate the management of emails increases for several reasons, such as spam e-mail classification, phishing e-mail classification, and multi-folder categorization, among others. To achieve the objective of study, analysis and comprehensive review to explore the classification as per the importance of emails as users need to look at these mails. The main area of classification is the primary inbox which contains some of the very important mails so to prioritize these mails we are using Natural Language Processing, here we are removing the stop-words present in it and assigning weights to the remaining words. By using frequency count, weight, and access time we can prioritize the mails so that it will be efficient for users to look for the important mails only. The research directions, research challenges, and open issues in the field of e-mail classification are also presented for future researchers.

The liver is one of the most significant organs in the human body. We can predict liver disease in a patient at an early stage based on previously predicted values using data from patients with abnormal liver function. Which helps the doctors to make a diagnosis. In this paper, the liver function test is analyzed for predicting liver disease, where the input of the patient's details and output data are passed into various classifiers such as Support Vector Machine, K-Nearest Neighbor, Hard Voting Classifier, and Deep Neural Network Multilayer Perceptron Techniques. Model Evaluation Criteria such as the Confusion Matrix, Precision Score, Recall, Accuracy, Specificity, and F-score are used to determine the best model. A dataset of 583 individuals suffering from liver disease is analyzed and we found that Hard Voting Classifier (HVC) is the best for this dataset. Additionally, this Voter Classifier prediction algorithm gives higher accuracy, which will help to diagnose liver disease. INDEX TERMS Feed forward network, perceptron algorithm, support vector machine, voting classifier.

Learning outcomes are measurable statements that articulate educational aims in terms of what knowledge, skills, and other competences students possess after successfully completing a given learning experience. This paper presents an analysis of the disparity between the claimed and formulated learning outcomes categorized in knowledge, skills, and social responsibility competency classes as it is postulated in the European Qualification Framework. We employed machine learning classification algorithms to detect and reveal main errors in their formulation that result in incorrect classification using generally available syllabus data from 22 universities. The proposed method was employed in two stages: preprocessing (creating a Python dataframe structure) and classification (by performing tokenization with the term frequency–inverse document frequency method). The obtained results demonstrated high effectiveness in correct classification for a number of machine learning algorithms. ...

The continuous growth of email users has resulted in the increasing of unsolicited emails also known as Spam. In current, server side and client side anti spam filters are introduced for detecting different features of spam emails. However, recently spammers introduced some effective tricks consisting of embedding spam contents into digital image, pdf and doc as attachment which can make ineffective to current techniques that is based on analysis digital text in the body and subject fields of email. Many of proposed working strategy provides an anti spam filtering approach that is based on data mining techniques which classify the spam and ham emails. The effectiveness of these approaches is evaluated on large corpus of simple text dataset as well as text embedded image dataset. But most of the filtering techniques are unable to handle frequent changing scenario of spam mails adopted by the spammers over the time. Therefore improved spam control algorithms or enhancing the efficiency of various existing data mining algorithms to its fullest extent are the utmost requirement. A comparative study is presented on various spam filtering techniques adopted on the basis of various attributes to find best among all to extract the best results.

Email being an efficient, cost-effective, real-time communication mode results into effective productivity among the professional in the organization. It constitutes almost 90% of daily office procedures in organizations, hence the productivity of organizations depends heavily on the text communicated in emails. The presented research work focuses on email profiling in organizations based on mail text interpretation and analysis. In the proposed work we will be working on datasets containing email communication of ENRON Corporation as test case. The profiling would be done using Text interpretation and analysis algorithm using machine learning algorithms. The BoW will be implemented to analyze and predict the characteristics of incoming and outgoing emails, then these could be mapped and profiled as per the behavior of employees into 3 categories of productive based on positive responses, neutral and non-productive based on negative responses.

Psychometric tests and personality assessments are widely used in a variety of settings, from academic research to employment screening. Traditional methods of administering and scoring these tests can be time-consuming and labor-intensive, making them difficult to scale or automate. However, recent advances in machine learning have made it possible to develop more efficient and accurate psychometric tests and personality assessments. This paper provides an overview of the use of machine learning in the development and administration of psychometric tests and personality assessments. We discuss the benefits and limitations of machine learning-based approaches, including the ability to process large amounts of data quickly and identify complex patterns that may not be immediately apparent to human observers. We also review some of the key challenges associated with using machine learning in this context, including the need for high-quality data and the potential for algorithmic bias. Overall, we argue that machine learning has the potential to revolutionize the field of psychometric testing and personality assessment, enabling researchers and practitioners to develop more efficient, accurate, and equitable assessment tools. However, we caution that careful consideration must be given to issues of data quality, bias, and fairness in order to ensure that machine learning-based assessments are truly effective and equitable I.

Stress is a prevalent issue that affects individuals' mental and physical well-being, leading to various health problems. The use of machine learning (ML) has been gaining popularity as a tool for stress detection. ML techniques have shown promising results in identifying patterns and features from various physiological and behavioral data sources such as heart rate, blood pressure, and speech signals. The primary goal of stress detection using ML is to provide accurate, non-invasive, and costeffective methods for early stress detection and intervention. Overall, stress detection using ML holds great promise in providing an objective, efficient, and scalable approach for stress monitoring and intervention. Further research is required to address the challenges associated with data collection, feature extraction, and model generalization to diverse populations and contexts.

Emails have become the most economical and fastest communication forms. However, during the past few years, the increment of email users has dramatically increased spam emails. Various anti-spam techniques have been developed to minimize if not eliminate the spam problem. In this paper, we study the disparity in the effectiveness of using different decision tree algorithms in email classification and combat spam problems. For that, we have chosen Universiti Utara Malaysia emails as a case study. To achieve the best possible classification accuracy, we compared all chosen algorithms’ performance, which are Random Forest, LMT, Decision Stump, J48, Random Tree, and REP Tree. The experimental results showed that the Decision Stump algorithm is more effective to be used in classifying the emails, and the F-measures, Precision, and recall score for the Decision Stump algorithm are higher than the other comparison algorithms.

Learning outcomes are measurable statements that articulate educational aims in terms of what knowledge, skills, and other competences students possess after successfully completing a given learning experience. This paper presents an analysis of the disparity between the claimed and formulated learning outcomes categorized in knowledge, skills, and social responsibility competency classes as it is postulated in the European Qualification Framework. We employed machine learning classification algorithms to detect and reveal main errors in their formulation that result in incorrect classification using generally available syllabus data from 22 universities. The proposed method was employed in two stages: preprocessing (creating a Python dataframe structure) and classification (by performing tokenization with the term frequency–inverse document frequency method). The obtained results demonstrated high effectiveness in correct classification for a number of machine learning algorithms. ...

The liver is one of the most significant organs in the human body. We can predict liver disease in a patient at an early stage based on previously predicted values using data from patients with abnormal liver function. Which helps the doctors to make a diagnosis. In this paper, the liver function test is analyzed for predicting liver disease, where the input of the patient's details and output data are passed into various classifiers such as Support Vector Machine, K-Nearest Neighbor, Hard Voting Classifier, and Deep Neural Network Multilayer Perceptron Techniques. Model Evaluation Criteria such as the Confusion Matrix, Precision Score, Recall, Accuracy, Specificity, and F-score are used to determine the best model. A dataset of 583 individuals suffering from liver disease is analyzed and we found that Hard Voting Classifier (HVC) is the best for this dataset. Additionally, this Voter Classifier prediction algorithm gives higher accuracy, which will help to diagnose liver disease. INDEX TERMS Feed forward network, perceptron algorithm, support vector machine, voting classifier.

Objective: To report a review of various machine learning and hybrid algorithms for detecting SMS spam messages and comparing them according to accuracy criterion. Data sources: Original articles written in English found in Sciencedirect.com, Google-scholar.com, Search.com, IEEE explorer, and the ACM library. Study selection: Those articles dealing with machine learning and hybrid approaches for SMS spam filtering. Data extraction: Many articles extracted by searching a predefined string and the outcome was reviewed by one author and checked by the second. The primary paper was reviewed and edited by the third author. Results: A total of 44 articles were selected which were concerned machine learning and hybrid methods for detecting SMS spam messages. 28 methods and algorithms were extracted from these papers and studied and finally 15 algorithms among them have been compared in one table according to their accuracy, strengths, and weaknesses in detecting spam messages of the Tiago ...

In this work we report on a pilot study where we used machine learning to predict whether students will correctly solve the classic "ballistic pendulum" problem based on an essay written by students elucidating their approach to solving the problem. Specifically, students were asked to describe the "principles, assumptions, and approximations" they used to solve the problem. Student essays were codified using the practices of natural language processing. Essays from two non-consecutive semesters were used for training/validation (N = 1441) and testing (N=1480). The final model used to make predictions was an ensemble classification scheme using random forest, eXtreme Gradient Boosting classifier (XGBoost), and logistic regression as estimators. Our accuracy in predicting students' correctness was around 80% with slightly higher accuracy in identifying students who incorrectly solved the problem and slightly lower in predicting student who correctly solved the problem.

Personal and business users prefer to use email 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. Finding out the important mails of the all the emails received on the same day is becoming difficult for many users, as the volume of critical emails continues to grow, the need to automate the management of emails increases for several reasons, such as spam e-mail classification, phishing e-mail classification, and multi-folder categorization, among others. To achieve the objective of study, analysis and comprehensive review to explore the classification as per the importance of emails as users need to look at these mails. The main area of classification is the primary inbox which contains some of the very important mails so to prioritize these mails we are using Natural Language Processing, here we are removing the stop-words present in it and assigning weights to the remaining words. By using frequency count, weight, and access time we can prioritize the mails so that it will be efficient for users to look for the important mails only. The research directions, research challenges, and open issues in the field of e-mail classification are also presented for future researchers.

The detection of masked face is becoming an essential part of health care safetydue to the pandemic caused by the coronavirus and the surveillance systems. One of the most challenging problems in face recognition systems is the accurate identification of faces in the presence of occlusion like wearing of glasses and masks. The current study proposes a novel convolutional neural network(CNN)-based model for accurate detection of faces in the presence of mask and glasses.The novel architecture of the model was developed using ten convolutional layers, five max-pooling layers, and a dropout layer. The Adam optimizer was used for optimization of the performance our model. Early stopping criteria in conjunction with the ReduceLROnPlateau class was employed to avoid the overfitting problem. Our proposed model could achieve the accuracy of 99.71% on the test dataset suggesting its superiorityto its existing counterparts. Based on the results, the suitability of the proposed model for face detection in the presence of occlution in real-life application has been recommended.

Big Data concern large-volume, growing data sets that are complex and have multiple autonomous sources. Earlier technologies were not able to handle storage and processing of huge data thus Big Data concept comes into existence. This is a tedious job for users to identify accurate data from huge unstructured data. So, there should be some mechanism which classify unstructured data into organized form which helps user to easily access required data. Classification techniques over big transactional database provide required data to the users from large datasets more simple way. There are two main classification techniques, supervised and unsupervised. In this paper we focused on to study of different supervised classification techniques. Further this paper shows application of each technique and their advantages and limitations.

Phishing is the act of attempting to steal a user's financial and personal information, such as credit card numbers and passwords by pretending to be a trustworthy participant, during online communication. Attackers may direct the users to a fake website that could seem legitimate, and then gather useful and confidential information using that site. In order to protect users from Social Engineering techniques such as phishing, various measures have been developed, including improvement of Technical Security. In this paper, we propose a new technique, namely, "A Prediction Model for the Detection of Phishing e-mails using Topic Modelling, Named Entity Recognition and Image Processing". The features extracted are Topic Modelling features, Named Entity features and Structural features. A multi-classifier prediction model is used to detect the phishing mails. Experimental results show that the multi-classification technique outperforms the single-classifier-based prediction techniques. The resultant accuracy of the detection of phishing e-mail is 99% with the highest False Positive Rate being 2.1%.

Leukemia is type of cancer in blood which impacts the lymphatic framework and the bone marrow and also impacts white blood cells. Leukemia, in contrast to other types of cancer, does not produce solid tumors; instead, it produces a huge number of aberrant white blood cells that crowd out the healthy blood cells. Machine learning algorithms that are widely utilized in the treatment of leukemia, whether it is to classify the various forms of leukemia or to determine whether a patient has the disease. It is a malignant kind of cancer that results in a number of medical issues. Expert hematologists and pathologists manually examine blood samples under the microscope to make a diagnosis. Techniques like image processing and pattern recognition can be utilized to help these experts. In order to attain excellent performance in the categorization of malignant leukocytes challenge, this paper suggests straightforward modifications to conventional neural network topologies. Consequently, ther...

Internet of Things (IoT) involves a set of devices that aids in achieving a smart environment. Healthcare systems, which are IoT-oriented, provide monitoring services of patients' data and help take immediate steps in an emergency. Currently, machine learning-based techniques are adopted to ensure security and other non-functional requirements in smart health care systems. However, no attention is given to classifying the non-functional requirements from requirement documents. The manual process of classifying the non-functional requirements from documents is erroneous and laborious. Missing non-functional requirements in the Requirement Engineering (RE) phase results in IoT oriented healthcare system with compromised security and performance. In this research, an experiment is performed where non-functional requirements are classified from the IoT-oriented healthcare system's requirement document. The machine learning algorithms considered for classification are Logistic Re...

Under short messaging service (SMS) spam is understood the unsolicited or undesired messages received on mobile phones. These SMS spams constitute a veritable nuisance to the mobile subscribers. This marketing practice also worries service providers in view of the fact that it upsets their clients or even causes them lose subscribers. By way of mitigating this practice, researchers have proposed several solutions for the detection and filtering of SMS spams. In this paper, we present a review of the currently available methods, challenges, and future research directions on spam detection techniques, filtering, and mitigation of mobile SMS spams. The existing research literature is critically reviewed and analyzed. The most popular techniques for SMS spam detection, filtering, and mitigation are compared, including the used data sets, their findings, and limitations, and the future research directions are discussed. This review is designed to assist expert researchers to identify open areas that need further improvement.

With the most preferred communication method e-mails have become part of day to day life. Spams which are also called unwanted, junk ,unsolicited mail is one of the major problem in using the e-mails. There are basically two things that are confused with each other that are one is Paper junk mail and spam mail, these two are usually confused

India as we know is a densely populated country and Every year more than 6 crores of Indians graduates from diverse backgrounds and with diversity in education. Almost similar number of students enter into colleges for taking various education to help them in seeking jobs. Many sectors have experienced tremendous growth in employment and thus masses opt for those sectors whereas in many sectors there is huge unemployment either due to low jobs availability or demand of skilled workers is required. Thinking of the each and every branch and when comparing it with the current employment in India and abroad, we will definitely find some points that will help in predicting the admissions and jobs scenarios in the fields of engineering and technology, management and pharmacy. Due to the changing technology and its requirement for getting employed in India and abroad, there has to improvements suggested by experts for predicting the Prediction of Admission & Jobs in Engineering & Technology with respect to demographic locations. This is not a one time process and needs to be done frequently as trends in the industry keep changing. Addressing this problem will introduce the required changes that would bring the current youth and upcoming generations in parallel with the students of other countries in terms of knowledge and skills in that domain. There is a need to forecast the current trend in the admissions and job sectors so as to blend the courses and syllabus accordingly to keep the youth employed and skilled with rapidly changing world. Here we will achieve it by using Machine Learning algorithm. I. INTRODUCTION A. Motivation India, as we all know, is a highly populated nation, and every year, more than 6 crore Indians graduate from a variety of educational backgrounds and socioeconomic backgrounds. A nearly equal percentage of students enroll in institutions to pursue various degrees that will aid in their career search. In comparison to ten years ago, the demands of the work market have changed significantly with the advancement of technology. While there is significant unemployment in many industries due to either a lack of jobs or a need for qualified workers, several sectors have seen remarkable development in employment and as a result, many people choose to work in those sectors. When considering each and every branch and contrasting it with the employment trends both in India and abroad, we will undoubtedly discover some factors that will assist in predicting admissions and employment scenarios in the fields of management, pharmacy, and engineering and technology. B. Problem Statement Improvements have been suggested by experts for predicting the Prediction of Admission & Jobs in Engineering & Technology, Management, and Pharmacy with respect to demographic locations due to the changing technology and its requirement for employment in India and abroad. Due to the constant shift in industry trends, this is not a one-time process that needs to be completed frequently. If this problem is solved, the necessary changes will be made, putting the young people of today and the generations to come up with the same knowledge and skills as students in other countries. However, despite AICTE's approval of the proposed courses, graduates of such institutions, engineering divisions, management departments, schools, and others continue to be unemployed. There is currently no method for estimating or forecasting the short-and long-term employment potential of any Engineering, Management, Pharmacy, or other course by combining data from various sources and developing a computer program or application. In order to keep young people employed and skilled in a world that is rapidly changing, it is necessary to forecast the current trend in the admissions and employment sectors in order to blend the courses and syllabus accordingly.

Recent interest in the design of information retrieval systems that can balance an ability to find relevant content with an ability to protect sensitive content creates a need for test collections that are annotated for both relevance and sensitivity. This paper describes the development of such a test collection that is based on the Avocado Research Email Collection. Four people created search topics as a basis for assessing relevance, and two personas describing the sensitivities of representative (but fictional) content creators were created as a basis for assessing sensitivity. These personas were based on interviews with potential donors of historically significant email collections and with archivists who currently manage access to such collections. Two annotators then created relevance and sensitivity judgments for 65 topics, divided approximately equally between the two personas. Annotator agreement statistics indicate fairly good external reliability for both relevance and sensitivity annotations, and a baseline sensitivity classifier trained and evaluated using cross-validation achieved better than 80% 1 , suggesting that the resulting collection will likely be useful as a basis for comparing alternative retrieval systems that seek to balance relevance and sensitivity. CCS CONCEPTS • Information systems → Test collections.

India as we know is a densely populated country and Every year more than 6 crores of Indians graduates from diverse backgrounds and with diversity in education. Almost similar number of students enter into colleges for taking various education to help them in seeking jobs. Many sectors have experienced tremendous growth in employment and thus masses opt for those sectors whereas in many sectors there is huge unemployment either due to low jobs availability or demand of skilled workers is required. Thinking of the each and every branch and when comparing it with the current employment in India and abroad, we will definitely find some points that will help in predicting the admissions and jobs scenarios in the fields of engineering and technology, management and pharmacy. Due to the changing technology and its requirement for getting employed in India and abroad, there has to improvements suggested by experts for predicting the Prediction of Admission & Jobs in Engineering & Technology /Management/Pharmacy with respect to demographic locations. This is not a one time process and needs to be done frequently as trends in the industry keep changing. Addressing this problem will introduce the required changes that would bring the current youth and upcoming generations in parallel with the students of other countries in terms of knowledge and skills in that domain. There is a need to forecast the current trend in the admissions and job sectors so as to blend the courses and syllabus accordingly to keep the youth employed and skilled with rapidly changing world. Here we will achieve it by using MaAhine Learning algorithm. I.

Most of the public spaces are being used for the development of public buildings in the name of schools, hospitals, multipurpose halls, temples, community buildings, etc. The lack of lively and quality public places is faced by all in recent years. There are several reasons why our city continues to encroach on public spaces. The public, even the city leaders are not well aware of the concept and importance of public space. A lack of capacity for development activities and a lack of competent human resources still exist in the establishment and management of public space. Political will and interest play vital roles in achieving lively and vibrant public spaces. The total area or percentage composition of public open spaces to the total area of the municipality illustrates how much people-friendly environment the municipality has been able to provide to its citizens. The aim of the research (which is based on the field survey of 2019 A.D.) is to identify the public open spaces in Damak municipality, assess their use and potentiality; and formulate policies regarding the management of these spaces. A combined strategies technique was considered for the facts series, which includes a self-administered questionnaire survey and observations. The research has identified 167 public spaces in this municipality and it concludes that the formulation of guidelines and laws, with their proper implementation, is a must to secure safe public spaces.

The sixth-generation (6G) has stricter criteria for the online learning capability and high interpretability of taught algorithms. It is anticipated that machine learning would be crucial for making network effective and flexible, however the most promising technologies are frequently considered as secret elements because of their profound designs’ major areas of strength for and. To make AI calculations more reasonable to 6G-empowered web of things (IoT) organizations, the motivation behind this paper is to analyse their translations. This article presents two different ways for acquiring translations: the free technique and the Joint strategy. Probes numerous IoT network datasets exhibit that the recommended techniques produce unrivalled execution regarding the two clarifications and forecasts.

Binary classification is challenging when dealing with imbalanced data sets where the important class is made of extremely rare events, usually with a prevalence of around 0.1%. Such data sets are common in various real-world problems. Examples are medical diagnostics where the disease is considered a rare event, or multiple types of fraud detection where regular transactions are the most prevalent. The events are categorized as either predicted positive or predicted negative against a certain threshold. In large imbalanced data sets, it is expensive to verify, through human raters, all the predictions made by the classifier. As a result, only predicted positive events are rated. Additionally, in most industrial applications, it is useful to combine multiple classifiers to make a decision. We developed a machine learning pipeline which, combined with expert knowledge, decides on an optimal threshold. ROC curves are reformulated to true positive (TP) versus false positive (FP) curves...

Spam and phishing emails are not only annoying to users, but are a real threat to internet communication and web economy. The fight against unwanted emails has become a cat-and-mouse game between criminals and people trying to develop techniques for detecting such unwanted emails. Criminals are constantly developing new tricks and adopt the ones that make emails pass spam filters. We have developed a systems that enables the detection of certain common salting tricks that are employed by criminals. Salting is the intentional addition or distortion of content. In this paper we describe a framework to identify email messages that might contain new, previously unseen tricks. To this end, we compare the simulated perceived email message text generated by our hidden salting simulation system to the OCRed text we obtain from the rendered email message. We present robust text comparison techniques and train a classifier based on the differences of these two texts. In simulations we show that we can detect suspicious emails with a high level of accuracy.

Emails have become one of the major applications in daily life. The continuous growth in the number of email users has led to a massive increase of unsolicited emails, which are also known as spam emails. Managing and classifying this huge number of emails is an important challenge. Most of the approaches introduced to solve this problem handled the high dimensionality of emails by using syntactic feature selection. In this paper, an efficient email filtering approach based on semantic methods is addressed. The proposed approach employs the WordNet ontology and applies different semantic based methods and similarity measures for reducing the huge number of extracted textual features, and hence the space and time complexities are reduced. Moreover, to get the minimal optimal features' set, feature dimensionality reduction has been integrated using feature selection techniques such as the Principal Component Analysis (PCA) and the Correlation Feature Selection (CFS). Experimental results on the standard benchmark Enron Dataset showed that the proposed semantic filtering approach combined with the feature selection achieves high computational performance at high space and time reduction rates. A comparative study for several classification algorithms indicated that the Logistic Regression achieves the highest accuracy compared to Naïve Bayes, Support Vector Machine, J48, Random Forest, and radial basis function networks. By integrating the CFS feature selection technique, the average recorded accuracy for the all used algorithms is above 90%, with more than 90% feature reduction. Besides, the conducted experiments showed that the proposed work has a highly significant performance with higher accuracy and less time compared to other related works.

Timely analysis of cyber-security information necessitates automated information extraction from unstructured text. While state-of-the-art extraction methods produce extremely accurate results, they require ample training data, which is generally unavailable for specialized applications, such as detecting security related entities; moreover, manual annotation of corpora is very costly and often not a viable solution. In response, we develop a very precise method to automatically label text from several data sources by leveraging related, domain-specific, structured data and provide public access to a corpus annotated with cyber-security entities. Next, we implement a Maximum Entropy Model trained with the average perceptron on a portion of our corpus (∼750,000 words) and achieve near perfect precision, recall, and accuracy, with training times under 17 seconds.

Email spam has become a prevalent issue in recent times, with the growing number of internet users, spam emails are also on the rise. Many individuals use them for illegal and unethical activities such as phishing and fraud. Spammers send dangerous links through spam emails, which can harm our systems and gain access to personal information. It has become easier for criminals to create fake profiles and email accounts. They often impersonate real individuals in their spam emails, making them difficult to identify. This project aims to identify and detect fraudulent spam messages. The paper will explore the use of machine learning techniques, algorithms, and apply them to data sets. The goal is to select the best methods for maximum precision and accuracy in email spam detection.

In the virtual world, many internet applications are used by a mass of people for several purposes. Internet applications are the basic needs of people in the modern days of lifestyle which are also making habitual society. Like social media, e-mail technology is also more prevalent among people of different categories for personal and official communications. The widespread use of e-mailbased communication is also raising various types of cybercrimes, including cyberstalking. Cyberstalkers also use an e-mail-based approach to harass the victim in the form of cyberstalking. Cyberstalkers utilize several content-wise and intent-wise approaches to target the victim, such as spamming, phishing, spoofing, malicious, defamatory, e-mail bombing, and non-spam e-mails, including sexism, racism, and threatening, and finally, trying to hack the account over e-mail technology. This paper proposed an EBCD model for automatic cyberstalking detection on textual data of e-mail using the multi-model soft voting technique of the machine learning approach. Initially, experimental works were performed to train, test, and validate all classifiers of three model sets on three different labeled datasets. Dataset D1 contains spam, fraudulent, and phishing e-mail subject, dataset D2 contains spam e-mail body text, while dataset D3 contains harassment-related data. After that, trained, tested, and validated classifiers of all model sets were applied as a combined approach to automatically classify the unlabeled e-mails from the user's mailbox using the multi-model soft voting technique. The proposed EBCD model successfully classifies the e-mails from the user's mailbox into cyberstalking e-mails, suspicious e-mails (spam and fraudulent), and normal e-mails. In each model set of the EBCD model, several classifiers, namely support vector machine, random forest, naïve bayes, logistic regression, and soft voting, were used. The final decision in classifying the e-mails from the user's mailbox was taken by the soft voting technique of each model set. The TF-IDF feature extraction method was used with the entire applied machine learning model sets to obtain the feature vectors from the data. Experimental results show that the soft voting technique not only enhances the performance of the e-mail classification task but also supports making the right decision to avoid the wrong classification. Overall performance of the soft voting technique was better than other classifiers, although the performance of the support vector machine was also notable. As per experimental results, the soft voting technique obtained an accuracy of 97.7%, 97.7%, 98.9%, a precision of 97%, 98.3%, 98.6%, recall of 98.3%, 96.5%, 99.1%, f-score of 97.6%, 97.4%, 98.9%, and AUC of 99.4%, 99.7%, 99.9% on dataset D1, D2, and D3 respectively. The average performance of soft voting of each model set on classified e-mails from the user's mailbox was also notable, with an accuracy of 96.3%, precision of 98.1%, recall of 94%, f-score of 95.9%, and AUC of 96.8%.

Human beings have various capability that is helpful for acquiring different types of knowledge form environment or surroundings. In which of them to find human gender by their voice is an easy task to the human because human as a growing phase of life acquires knowledge form surrounding and listening data. But when it comes to the computer, then it becomes a difficult task to identify the gender of the human being by their voice. To obtain this task from the machine or robots we need to train that particular machine by providing relevant features, methodologies, and diverse training dataset. In this research paper, we proposed a model by combining the Naïve Bayes and deep learning methods for gender identification. In this work, we consider the acoustic features of voice to classify the males and females speech. To persuade this work, we collect around 2000 voice sample form different sentence acoustic from classmates and different website voice samples and used that voice samples as an input dataset form our proposed model. The duration of voice in dataset is greater than 3 seconds to less than 8 seconds. For the model training and testing we used dataset in the ratio of one-third (66% samples for training and 33% samples for testing). Dataset have 50-50% portion for both gender voices. The proposed model is based on two classifier name as Neural Network classifier and Naïve Bayes classifier. These two classifier gives prominent result with this dataset. The accuracy of naïve Bayes is 98% and neural Network gives 99%.

Email categorization is a critical function in any email client since it allows you to manage and arrange your emails into semantic groupings. Following the success of statistical artificial intelligence and machine learning in many areas of information management, these methods have become a standard strategy for email categorization. Using multi-labelled emails, this article offers a method of classifying them and recommending additional labels to users. Under the labels used by the user's contacts, new labels are recommended. Not only would the labels be recommended, but it would also sort the new emails into the categories according to those labels. The Random Forest Classifier is more effective in all kinds of email categorization than other algorithms. As a result, we would base our primary method on RFC with a few minor modifications made. To forecast the categories of new emails, we would use the user's past emails and categories. When Nave Bayes applies to the testing dataset, the average recall is 63%, and the Random Forest Classifier is used, the average recall is 64 %, respectively.

The aim of malware analysis is to detect whether a file is infected or not in order to avoid any kind of system intrusion. The goal of this research is to find the optimal machine learning algorithm to predict whether a file is malicious or not by using different machine learning models. For the above purpose the implementation and accuracy comparisons are done with the help of python libraries and then summary analysis is then used to suggest the best machine learning model for the detection of the malware infected files this can then be used as a layer in a bigger neural network for dynamic malware analysis and attack detection and prevention. The paper also highlights a few very important file characteristics that could possibly be used for malware detection.

The vehicular adhoc network (VANET), a developing study area in the intelligent transportation system, provides the network's vehicles with crucial information. Road accidents harm around 160 000 people;thus, they must be reduced, and VANET safety must be improved. The ability to foresee traffic jams is crucial for reducing vehicle accidents and enhancing pedestrian traffic flow. However, deep learning models' performance in foretelling traffic congestion on roadways is hampered by the network's dynamic behaviour of the vehicles.The utilization of AI in 6G vehicular organizations to help vehicular application administrations is a hotly debated issue for the latest examination works in the writing and is at present the subject of broad review. This paper offers a careful investigation of past examinations that consolidated support and profound support learning calculations for the administration of vehicular organizations, with an emphasis on vehicular broadcast communications troubles. Then, we go through RL and, in particular, DRL techniques to deal with new problems in 6G vehicular networks. Finally, we talk about and point out certain issues that need more research.

In Renewable energy schemes, Solar photovoltaic (PV) systems provide effective incorporation of generating electrical energy. Many current control techniques such as Hysteresis control, predictive control and Sliding mode control are available to improve the performance of PV systems. However, the current tracking of the existing controllers is suffered when connected to the grid-connected system due to the lack of constant switching frequency in the three-phase inverter control. Furthermore, the complexity of these systems is very high. An Artificial Neural Network (ANN) based Sliding Mode Control approach has been used to solve the voltage regulation in the gridconnected solar system. This model has been employed to implement a two-loop controller using a voltage controller as an outer loop and a current controller as an inner loop. The ANN-SMC-based three-phase inverter is simulated to diminish the load current's harmonics and provide robustness in the inverter control. The proposed system results show the robustness improvement of the gridconnected solar system with a low level of Total Harmonic Distortion (THD).

In many nations, the prevalence of diabetes is rising, and its impact on national health cannot be overlooked. Smart medicine is a medical concept in which technology is used to aid in disease detection and treatment. The objective of this study is to take a gander 13946 at the information and look at changed diabetic mellitus forecasting algorithms. According to rising dismalness as of late, the quantity of diabetic patients worldwide will arrive at 642 million out of 2040, suggesting that one out of each 10persons would be affected. This worrisome figure, without a question, demands immediate attention. AI has been applied to an assortment of aspects of clinical wellbeing as a result of its rapid progress. To predict diabetes mellitus in this review, we utilized a choice tree, an arbitrary timberland, and a neural organization.

Sp-DELES scale; and the second to measure academic satisfaction using the SA scale. The results demonstrated a perfect positive relationship with a = .935. determining a relationship between the psychosocial learning environment of distance education and the academic satisfaction of graduate students during the pandemic.

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.

Email categorization is a critical function in any email client since it allows you to manage and arrange your emails into semantic groupings. Following the success of statistical artificial intelligence and machine learning in many areas of information management, these methods have become a standard strategy for email categorization. Using multi-labelled emails, this article offers a method of classifying them and recommending additional labels to users. Under the labels used by the user's contacts, new labels are recommended. Not only would the labels be recommended, but it would also sort the new emails into the categories according to those labels. The Random Forest Classifier is more effective in all kinds of email categorization than other algorithms. As a result, we would base our primary method on RFC with a few minor modifications made. To forecast the categories of new emails, we would use the user's past emails and categories. When Nave Bayes applies to the testing dataset, the average recall is 63%, and the Random Forest Classifier is used, the average recall is 64 %, respectively.

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 emails 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.

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.

Big Data concern large-volume, growing data sets that are complex and have multiple autonomous sources. Earlier technologies were not able to handle storage and processing of huge data thus Big Data concept comes into existence. This is a tedious job for users unstructured data. So, there should be some mechanism which classify unstructured data into organized form which helps user to easily access required data. Classification techniques over big transactional database provide required data to the users from large datasets more simple way. There are two main classification techniques, supervised and unsupervised. In this paper we focused on to study of different supervised classification techniques. Further this paper shows a advantages and limitations.