Pawan Kumar Verma
@sharda.ac.in
Associate Professor, Computer Science & Engineering Department
Sharda University, Greater Noida, Uttar Pradesh, India
Associate Professor(CSE) |ML Researcher |Mentor@ Worldwide AI Hackathon |Mentor@ upGrad | 1 IEEE Transactions | 6 SCIE | 6 Scopus | 3 ESCI | Wipro Certified Java Trainer | 11.5 Yrs Exp. | Pega CSA Certified
RESEARCH, TEACHING, or OTHER INTERESTS
Artificial Intelligence
Scopus Publications
Scholar Citations
Scholar h-index
Scholar i10-index
Scopus Publications
Ajeet Kumar Sharma, Anushree, Nitin Rakesh, and Pawan Kumar Verma
IEEE
The persistent and evolving threat of phishing attacks demands effective and adaptive detection techniques. This research paper presents a comprehensive evaluation and comparison of various machine learning approaches to detect phishing attacks. We investigated five prominent algorithms: Logistic Regression, Support Vector Machine (SVM), K-Nearest Neighbors (K-NN), Naive Bayes, and Extreme Gradient Boosting (XGBoost), to determine their efficacy in identifying phishing activities. Our methodology involved a systematic analysis using a large dataset of phishing and legitimate URLs, where each model was trained, tested, and validated to ensure robustness and reliability. The performance of each algorithm was assessed based on accuracy, precision, recall, and F1 score. Among the evaluated models, XGBoost demonstrated superior performance, achieving an exceptional accuracy of 99.75%. This result underscores the potential of XGBoost in phishing attack detection, offering a promising tool for cybersecurity applications.
Pawan Kumar Verma, Prateek Agrawal, Vishu Madaan, and Radu Prodan
Springer Science and Business Media LLC
AbstractOnline social media enables low cost, easy access, rapid propagation, and easy communication of information, including spreading low-quality fake news. Fake news has become a huge threat to every sector in society, and resulting in decrements in the trust quotient for media and leading the audience into bewilderment. In this paper, we proposed a new framework calledMessageCredibility (MCred) for fake news detection that utilizes the benefits of local and global text semantics. This framework is the fusion of Bidirectional Encoder Representations from Transformers (BERT) using the relationship between words in sentences for global text semantics, and Convolutional Neural Networks (CNN) using N-gram features for local text semantics. We demonstrate through experimental results a popular Kaggle dataset that MCred improves the accuracy over a state-of-the-art model by 1.10% thanks to its combination of local and global text semantics.
Pawan Kumar Verma, Prateek Agrawal, Vishu Madaan, and Charu Gupta
Springer Science and Business Media LLC
Ajitesh Kumar, Akhilesh Kumar Singh, Ijaz Ahmad, Pradeep Kumar Singh, Anushree, Pawan Kumar Verma, Khalid A. Alissa, Mohit Bajaj, Ateeq Ur Rehman, and Elsayed Tag-Eldin
MDPI AG
Nowadays, in a world full of uncertainties and the threat of digital and cyber-attacks, blockchain technology is one of the major critical developments playing a vital role in the creative professional world. Along with energy, finance, governance, etc., the healthcare sector is one of the most prominent areas where blockchain technology is being used. We all are aware that data constitute our wealth and our currency; vulnerability and security become even more significant and a vital point of concern for healthcare. Recent cyberattacks have raised the questions of planning, requirement, and implementation to develop more cyber-secure models. This paper is based on a blockchain that classifies network participants into clusters and preserves a single copy of the blockchain for every cluster. The paper introduces a novel blockchain mechanism for secure healthcare sector data management, which reduces the communicational and computational overhead costs compared to the existing bitcoin network and the lightweight blockchain architecture. The paper also discusses how the proposed design can be utilized to address the recognized threats. The experimental results show that, as the number of nodes rises, the suggested architecture speeds up ledger updates by 63% and reduces network traffic by 10 times.
Ashutosh Shankhdhar, Pawan Kumar Verma, Prateek Agrawal, Vishu Madaan, and Charu Gupta
Emerald
PurposeThe aim of this paper is to explore the brain–computer interface (BCI) as a methodology for generating awareness and increasing reliable use cases of the same so that an individual's quality of life can be enhanced via neuroscience and neural networks, and risk evaluation of certain experiments of BCI can be conducted in a proactive manner.Design/methodology/approachThis paper puts forward an efficient approach for an existing BCI device, which can enhance the performance of an electroencephalography (EEG) signal classifier in a composite multiclass problem and investigates the effects of sampling rate on feature extraction and multiple channels on the accuracy of a complex multiclass EEG signal. A one-dimensional convolutional neural network architecture is used to further classify and improve the quality of the EEG signals, and other algorithms are applied to test their variability. The paper further also dwells upon the combination of internet of things multimedia technology to be integrated with a customized design BCI network based on a conventionally used system known as the message query telemetry transport.FindingsAt the end of our implementation stage, 98% accuracy was achieved in a binary classification problem of classifying digit and non-digit stimuli, and 36% accuracy was observed in the classification of signals resulting from stimuli of digits 0 to 9.Originality/valueBCI, also known as the neural-control interface, is a device that helps a user reliably interact with a computer using only his/her brain activity, which is measured usually via EEG. An EEG machine is a quality device used for observing the neural activity and electric signals generated in certain parts of the human brain, which in turn can help us in studying the different core components of the human brain and how it functions to improve the quality of human life in general.
Pawan Kumar Verma and Prateek Agrawal
Springer Nature Singapore
Pawan Kumar Verma, Prateek Agrawal, Ivone Amorim, and Radu Prodan
Institute of Electrical and Electronics Engineers (IEEE)
Social media is a popular medium for the dissemination of real-time news all over the world. Easy and quick information proliferation is one of the reasons for its popularity. An extensive number of users with different age groups, gender, and societal beliefs are engaged in social media websites. Despite these favorable aspects, a significant disadvantage comes in the form of fake news, as people usually read and share information without caring about its genuineness. Therefore, it is imperative to research methods for the authentication of news. To address this issue, this article proposes a two-phase benchmark model named WELFake based on word embedding (WE) over linguistic features for fake news detection using machine learning classification. The first phase preprocesses the data set and validates the veracity of news content by using linguistic features. The second phase merges the linguistic feature sets with WE and applies voting classification. To validate its approach, this article also carefully designs a novel WELFake data set with approximately 72 000 articles, which incorporates different data sets to generate an unbiased classification output. Experimental results show that the WELFake model categorizes the news in real and fake with a 96.73% which improves the overall accuracy by 1.31% compared to bidirectional encoder representations from transformer (BERT) and 4.25% compared to convolutional neural network (CNN) models. Our frequency-based and focused analyzing writing patterns model outperforms predictive-based related works implemented using the Word2vec WE method by up to 1.73%.
Pawan Kumar Verma and Prateek Agrawal
Springer Singapore
Pawan Kumar Verma, Vivek Sharma, and Shalini Agarwal
IEEE
Fake news has been evolved since the internet use has been increased and mainly when the focus has been diverted on to some social networking sites for gaining any kind of information or news. The widely accepted definition of fake news is the news that is not true or unreal spreaded either intentionally or unintentionally. people are using twitter, Facebook and Instagram for seeking any kind of news or information. In this paper we have proposed the framework to find out the authenticity of the twitter user and tweets by computing tweet score and user score. With the help of these scores we will be able to label the tweet and its user as an authentic tweet or user that has been verified by some methods.
Vivek Sharma, Shalini Agarwal, and Pawan Kumar Verma
IEEE
Differential Evolution algorithm has proved to be effective and best method for solving various optimization challenges. It has been proved to be rather cumbersome to manually set control parameters in DE. This paper sketch a new variant of the DE algorithm that provides an environment to auto-adjust the control parameters settings. For the past years, DE has captured the attention in many practical cases. It makes use of a few control parameters that are bound to the same value throughout the evolutionary process. Manual control parameters setting is a time-consuming process, so the proposed work provides a reliable, accurate and fast technique to optimize numerical function. This work is tested against various numerical set functions. Final results show that this proposed algorithm performs a cut above when compared with the classical Differential Evolution algorithm, and the other control parameter setting variant of DE considered in the literature.
Shalini Agarwal, Vivek Sharma, and Pawan Kumar Verma
IEEE
Nowadays, Palm Print is one of the most reliable biometric traits among all of personal identification due to its high stability and various unique as well as stable features like principal lines, minutiae points, singular points, ridges, textures etc. In this paper we present a new method using Curvelet energy distribution algorithm. We extract texture feature from palm image using Curvelet Energy Distribution Algorithm (CEDA), in which First, 2 level curvelet transform is performed over palm image and then energy distribution of each sub-bands will be calculated at both level of transform. These energy distributions will be collected as feature vector, sort them and use as a texture feature of an image. This feature vector generated after sorting does not change for input palm image if one person scans his palm in any angle hence effectively achieve good recognition rate with rotation invariant property. Multi class SVM classifier is used for classification which is a less complex high performance classifier. Experiment will be performed over PolyU palm print database collected over 250 persons. This proposed algorithm compared with recognition using wavelet transform and texture extraction using Gabor filter and achieve a good recognition rate.
Pawan Kumar Verma, Shalini Agarwal, and Mohd Aamir Khan
IEEE
Today's E-commerce is totally based upon the opinions of the customers. There are lots of opinions on the internet about any product. Now a day's people are using roman words to express their reviews. The main aim of this paper is to find the polarity of any item. In this paper firstly Roman opinion words are converted into corresponding English opinion words then applied Jaccard Similarity algorithm based on clustering for obtaining the polarity of the item. This paper approaches to the conclusion of the overall polarity of the item efficiently.
Shalini Agarwal, Pawan Kumar Verma, and Mohd Aamir Khan
IEEE
Identity verification of a person always considered a wide research area. In past, Identity of a person verified by traditional token-based or knowledge-based system but in recent year's bio-metrics traits like face, finger, iris, palm print etc. become a key technology for identity verification. Palm print is also the one bio-metric trait that can be used for the efficient identification. Palm print contains many unique features like principal lines, points, ridges, textures etc. that can differentiate two person. Because palm is a large area of hand, there is a common problem of palm displacement over scanner that results in increase in false rejection rate. This paper proposed a method which first generates ROI of captured palm image then median filtering is applied to remove noise and increasing edge sharpness. Histogram equalization applied after that for contrast stretching for low resolution images. Enhanced image is then divided in sixteen equal part, texture feature is extracted from each part of image separately using different orientations of Gabor filter. The generated feature vectors of all sixteen images are then normalized to a single feature vector using n bin histogram process. This increase acceptance rate in case if palm is placed over scanner in slightly different angles because working on small areas of palm helps to extract detail features. This paper used SVM for classification of generated feature vector and Experiment performed on polyU palm print database [1].
Pawan Kumar Verma, Deepak Dahiya, and Pooja Jain
IEEE
A typical software system comprises of several crosscutting concerns (also known as aspects). Code tangling and scattering are two difficulties that occur in current software implementation methodologies which affect software design and development in many ways like, poor traceability, lower productivity, less code reuse and poor code quality. Aspect Oriented Programming (AOP), which allows for modularizing of concerns that normally cause crosscutting in object oriented system, has efficiently solved the problem that the Object Oriented Programming has encountered such as the scattered codes and tangled codes resulting from the cross cutting concerns. Aspect-Oriented Software Development (AOSD) is becoming a new technique, which provides modularization of crosscutting concerns. The aim of this paper is to define an Aspect Oriented Software Architecture for software development with minimum code tangling and scattering. By this architecture not only the design efficiency can be improved but also the model built is easier to comprehend and reuse.
Pawan Kumar Verma, Neetu Singh, and Rahul Katarya
IEEE
The creation of efficient and secure framework is becoming important with the rapid development of internet and World Wide We b (WWW). The objective of this paper is to examine the web based framework to retain the information of any organization. Thus, we propose a new scaffold on the basis of JSP Access Model and Model-View-Controller (MVC), which will assist in providing information regarding users and some indispensable services to several organizations. This framework also provides better security then other frameworks. In turn it supports faster retrieval and processing of data. For the implementation of this framework different approaches of software engineering are being used. In this paper work we discuss this framework with Java programming language, because Java is more popular and secure programming language for web a pplication.