To obtain a position that facilitates my interest in teaching and performance for contributing to the research works of related area - Information & Communication Technology [ICT].

Wireless Mobile Communication, Cognitive Radio Networking, Multimedia and Virtual Reality, Computer Networking, Computer Graphics & Information Systems

0

Extended the application area of Neural Network Computing in Biological data, Bio-Informatics, Text mining, Information Extraction, Pattern Recognition, Intelligence Communication Networking.

Ad-Hoc Networking, ATM Networking and Parallel Processing.

0

Wireless Communication, Integral Imaging, Image Processing, Machine Learning

Md Shariful Islam

This study presents a three-dimensioanl (3-D) Integral Imaging system to improve concurently the viewing angle, Depth of Focus, and Resolution of the reconstructed object images by using multiprojections and nonuniform lenslets array with Intermediate-View Reconstruction Method (IVRM). 

In this paper an attemppt has been made to compare three routing protocols on the basis of performance under different environments. The comparison has been done under the UDP payload (CBR traffic).

There are many routing protocols of these DSR, DSDV performance has been shown under different environment through this paper. 

 Through this paper one of the limitation of Integral Imaging three-dimensional (3D) display system is enhanced (Viewing angle. 

The main goal of this article is analyzing the recent work in multiple objects tracking to handle those challenges. 

This paper presents a Computer Generated Integral Imaging (CGII) system that makes better the visual quality of three-dimensioanl (3D) reconstructed images based on real depth of 3D object. 

This paper delas with a three-dimensional Integral Imaging systen to improve viewing angle by using multiple illuminations. 

In this paper we proposed a three-dimensional (3-D) Integral Imaging (II) system using double layer structure microlens array in place of conventional microlens array. 

The main goal of this work is to optimization of facial feature point by establishing a mathematical relationship among facial features and used optimize feature points for age classification. 

In this paper a system developed for speech encoding, analysis, synthesis and gender identification is presented 

The paper deals with the study based on the comparative analysis of radio propagation models for mobile cellular wireless communication of global system for mobile at frequencies 0.9 GHz and 1.8 GHz, respectively. The path loss propagation models arevital tool for planning the wireless network as well as designed to predict path loss in a meticulous environment. Various propagation models: Free-space model, CCIR (ITU-R) model, Hata model, Ericson model, and Stanford University Interim (SUI) model have been studied and examined through analytically from the base station (BS) to mobile station (MS) and vice versa followed by respective simulation performance evaluation by using Matlab simulator. The observed data is collected at the operating frequency of 0.9 GHz from various environments (high density region and low density region) using the spectrum analyzer and path loss comparison is shown for different model.

The prime aspect of this paper is to report the atmospheric impairment effects in wireless communication signalfor different frequencies. Comprehensive tests have been carried out to justify and confront theattenuation impacts of rain, cloud and water vapor oxygen for various wave length channels. The result of the atmospheric impairments is obtained by the parameters which are resembled to the real world measurements. Moreover, the performance is analyzed by the use of different propagation modules and transmission parameters that are frequently used in satellite and radar communications. Experimental resultsindicate that the attenuation effect due to rain has initiated at frequencies above 10 GHz while, other atmospheric effects preceded by attenuation of the cloud and gasses have less effect. It is also observed thatduring frequency transmission, the rate of rainfall and relative humidity become directly proportional to signal quality degradation.

test

I'm Professor Dr. Md. Sipon Miah (PhD, Ireland), Faculty member of Information & Communication Technology (ICT), Islamic University, Kushtia-7003, Bangladesh.

Wireless Communications, Wireless Sensor Networks, Cognitive Radio Networks, RF Energy Harvesting, Cloud Computing, Internet of Things, Machine Learning, etc.

0

https://scholar.google.com/citations?user=XhP8D5oAAAAJ&hl=en&oi=ao

Md. Sipon Miah (First author)

Md. Sipon Miah (First author)

Md. Sipon Miah (First author)

Md. Sipon Miah (First author)

Md. Sipon Miah (First author)

Md Sipon Miah (Co-author)

Abstract- The purpose of this paper is to investigate the performance of UDP over various routing protocols in ad hoc networks. For this investigation we have chosen three routing schemes, DSDV, DSR and AODV and four network scenarios of 4, 8, 16 and 32 nodes of various node mobility speeds. Results are produced by evaluating throughput and end to end packet delay over the UDP connection through simulation experiments.

Md. Sipon Miah (First author)

Absract- The cognitive radio relay plays a vital role in cognitive radio networking (CRN), as it can improve the cognitive sum rate, extend the coverage, and improve the spectral efficiency. However, cognitive relay aided CRNs cannot obtain a maximal sum rate, when the existing sensing approach is applied to a CRN. In this paper, we present an enhanced sum rate in the cluster based cognitive radio relay network utilizing a reporting framework in the sequential approach. In this approach a secondary user (SU) extends its sensing time until right before the beginning of its reporting time slot by utilizing the reporting framework. Secondly all the individual measurement results from each relay aided SU are passed on to the corresponding cluster head (CH) through a noisy reporting channel, while the CH with a soft-fusion report is forwarded to the fusion center that provides the final decision using the n-out-of-k-rule. With such extended sensing intervals and amplified reporting, a better sensing performance can be obtained than with a conventional non-sequential approach, therefore making it applicable for the future Internet of Things. In addition, the sum rate of the primary network and CCRRN are also investigated for the utilization reporting framework in the sequential approach with a relay using the n-out-of-k rule. By simulation, we show that the proposed sequential approach with a relay (Lemma 2) provides a significant sum rate gain compared to the conventional non-sequential approach with no relay (Lemma 1) under any condition.

Md. Sipon Miah (First author)

Absract- State-of-the-art energy detection (ED) based spectrum sensing requires perfect knowledge of noise power and is vulnerable to noise uncertainty. An eigenvalue-based spectrum sensing approach performs well in such an uncertain environment, but does not mitigate the spectrum scarcity problem, which evolves with the future Internet of Things (IoT) rollout. In this paper, we propose a multi-user multiple-input and multiple-output (MU-MIMO) based cognitive radio scheme for the Internet of Things (CR-IoT) with weighted-eigenvalue detection (WEVD) for the analysis of sensing, system throughput, energy efficiency and expected lifetime. In this scheme, each CR-IoT user is being equipped with MIMO antennas; we calculate the WEVD ratio, which is defined as the ratio between the difference of the maximum eigenvalue and minimum eigenvalue to the sum of the maximum eigenvalue and minimum eigenvalue. This mitigates against the spectrum scarcity problem, enhances system throughput, improves energy efficiency, prolongs expected lifetime and lowers error probability. Simulation results confirm the effectiveness of the proposed scheme; here the WEVD technique demonstrates a better detection gain and enhanced system throughput in comparison to the conventional scheme with eigenvalue based detection (EVD) and ED techniques in a noise uncertainty environment (i.e. SNR <-28). Furthermore, the proposed scheme has a lower energy consumption, prolonged expected lifetime and achieves a low error probability when compared with other schemes like the conventional single-input and single-output (SISO) based CR-IoT scheme with EVD and ED spectrum sensing.

Md. Sipon Miah (First author & Corresponding author)

Absract- Spectrum sensing plays a vital role in cognitive radio networks (CRNs) for identifying the spectrum hole. However, an individual cognitive radio user in a CRN does not obtain sufficient sensing performance and sum rate of the primary and secondary links to support the future Internet of Things (IoT) using conventional detection techniques such as the energy detection (ED) technique in a noise-uncertain environment. In an environment comprising noise uncertainty, the performance of conventional energy detection techniques is significantly degraded owing to the noise fluctuation caused by the noise temperature, interference, and filtering. To mitigate this problem, we present a cooperative spectrum sensing technique that comprises the use of the Kullback–Leibler divergence (KLD) in cognitive radio-based IoT (CR-IoT). In the proposed method, each unlicensed IoT device that is capable of spectrum sensing, which is called a CR-IoT user, makes a local decision using the KLD technique. The spectrum sensing performed with the KLD requires a smaller number of samples than other conventional approaches, e.g., energy detection, for reliable sensing even in a noise uncertain environment. After the local decision is made, each CR-IoT user sends its own local decision result to the corresponding fusion center, which makes a global decision using the soft fusion rule. The results obtained through simulations show that the proposed KLD scheme achieves a better sensing performance, i.e., higher detection and lower false-alarm probabilities, enhances the sum rate, and reduces the total time as compared to the conventional ED scheme under various fading channels.

Md. Sipon Miah* (Corresponding Author)

Absract- In this paper, we proposed the unscented Kalman filter (UKF) based on cooperative spectrum sensing (CSS) scheme in a cognitive radio network (CRN) using an adaptive fuzzy system—in this proposed scheme, firstly, the UKF to apply the nonlinear system which is used to minimize the mean square estimation error; secondly, an adaptive fuzzy logic rule based on an inference engine to estimate the local decisions to detect a licensed primary user (PU) that is applied at the fusion center (FC). After that, the FC makes a global decision by using a defuzzification procedure based on a proposed algorithm. Simulation results show that the proposed scheme achieved better detection gain than the conventional schemes like an equal gain combining (EGC) based soft fusion rule and a Kalman filter (KL) based soft fusion rule under any conditions. Moreover, the proposed scheme achieved the lowest global probability of error compared to both the conventional EGC and KF schemes.

Md. Sipon Miah* (Corresponding Author)

Abstract- In recent decades, human brain tumor detection has become one of the most challenging issues in medical science. In this paper, we propose a model that includes the template-based K means and improved fuzzy C means (TKFCM) algorithm for detecting human brain tumors in a magnetic resonance imaging (MRI) image. In this proposed algorithm, firstly, the template-based K-means algorithm is used to initialize segmentation significantly through the perfect selection of a template, based on gray-level intensity of image; secondly, the updated membership is determined by the distances from cluster centroid to cluster data points using the fuzzy C-means (FCM) algorithm while it contacts its best result, and finally, the improved FCM clustering algorithm is used for detecting tumor position by updating membership function that is obtained based on the different features of tumor image including Contrast, Energy, Dissimilarity, Homogeneity, Entropy, and Correlation. Simulation results show that the proposed algorithm achieves better detection of abnormal and normal tissues in the human brain under small detachment of gray-level intensity. In addition, this algorithm detects human brain tumors within a very short time—in seconds compared to minutes with other algorithms.

Md. Sipon Miah (Corresponding author)

Abstract- The Cognitive Radio based Internet of Things (CR-IoT) is a promising technology that provides IoT endpoints, i.e., CR-IoT users the capability to share the radio spectrum otherwise allocated to licensed Primary Users (PUs). Cooperative Spectrum Sensing (CSS) improves spectrum sensing accuracy in a CR-IoT network. However, its performance may be degraded by potential attacks of the malicious CR-IoT users that send their incorrect sensing information to the corresponding Fusion Center (FC). This study presents a promising Machine Learning (ML)-based malicious user detection scheme for a CR-IoT network that uses a Support Vector Machine (SVM) algorithm to identify and classify malicious CR-IoT users. The classification allows the FC to make a more robust global decision based on the sensing results (i.e., energy vectors) which are reported only by the normal CR-IoT users. The effectiveness of the proposed SVM algorithm based ML in a CR-IoT network with the malicious CR-IoT users is verified via simulations.

 Md. Sipon Miah (Co-author)

Md. Sipon Miah (Co-author)

Abstract- Skin cancer is one of the top three perilous types of cancer caused by damaged DNA that can cause death. This damaged DNA begins cells to grow uncontrollably and nowadays it is getting increased speedily. There exist some researches for the computerized analysis of malignancy in skin lesion images. However, analysis of these images is very challenging having some troublesome factors like light reflections from the skin surface, variations in color illumination, different shapes, and sizes of the lesions. As a result, evidential automatic recognition of skin cancer is valuable to build up the accuracy and proficiency of pathologists in the early stages. In this paper, we propose a deep convolutional neural network (DCNN) model based on deep learning approach for the accurate classification between benign and malignant skin lesions. In preprocessing we firstly, apply filter or kernel to remove noise and artifacts; secondly, normalize the input images and extract features that help for accurate classification; and finally, data augmentation increases the number of images that improves the accuracy of classification rate. To evaluate the performance of our proposed, DCNN model is compared with some transfer learning models such as AlexNet, ResNet, VGG-16, DenseNet, MobileNet, etc. The model is evaluated on the HAM10000 dataset and ultimately we obtained the highest 93.16% of training and 91.93% of testing accuracy respectively. The final outcomes of our proposed DCNN model define it as more reliable and robust when compared with existing transfer learning models.

Md. Sipon Miah (Co-author)

Abstract- The COVID-19 outbreak resulted in preventative measures and restrictions for Bangladesh during the summer of 2020—these unstable and stressful times led to multiple social problems (e.g., domestic violence and divorce). Globally, researchers, policymakers, governments, and civil societies have been concerned about the increase in domestic violence against women and children during the ongoing COVID-19 pandemic. In Bangladesh, domestic violence against women and children has increased during the COVID-19 pandemic. In this article, we investigated family violence among 511 families during the COVID-19 outbreak. Participants were given questionnaires to answer, for a period of over ten days; we predicted family violence using a machine learning-based model. To predict domestic violence from our data set, we applied random forest, logistic regression, and Naive Bayes machine learning algorithms to our model. We employed an oversampling strategy named the Synthetic Minority Oversampling Technique (SMOTE) and the chi-squared statistical test to, respectively, solve the imbalance problem and discover the feature importance of our data set. The performances of the machine learning algorithms were evaluated based on accuracy, precision, recall, and F-score criteria. Finally, the receiver operating characteristic (ROC) and confusion matrices were developed and analyzed for three algorithms. On average, our model, with the random forest, logistic regression, and Naive Bayes algorithms, predicted family violence with 77%, 69%, and 62% accuracy for our data set. The findings of this study indicate that domestic violence has increased and is highly related to two features: family income level during the COVID-19 pandemic and education level of the family members.

Md Sipon Miah (Corresponding Author)

Abstract- Spectrum sensing in a cognitive radio network involves detecting when a primary user vacates their licensed spectrum, to enable secondary users to broadcast on the same band. Accurately sensing the absence of the primary user ensures maximum utilization of the licensed spectrum and is fundamental to building effective cognitive radio networks. In this paper, we address the issues of enhancing sensing gain, average throughput, energy consumption, and network lifetime in a cognitive radio-based Internet of things (CR-IoT) network using the non-sequential approach. As a solution, we propose a Dempster–Shafer theory-based throughput analysis of an energy-efficient spectrum sensing scheme for a heterogeneous CR-IoT network using the sequential approach, which utilizes firstly the signal-to-noise ratio (SNR) to evaluate the degree of reliability and secondly the time slot of reporting to merge as a flexible time slot of sensing to more efficiently assess spectrum sensing. Before a global decision is made on the basis of both the soft decision fusion rule like the Dempster–Shafer theory and hard decision fusion rule like the “n-out-of-k” rule at the fusion center, a flexible time slot of sensing is added to adjust its measuring result. Using the proposed Dempster–Shafer theory, evidence is aggregated during the time slot of reporting and then a global decision is made at the fusion center. In addition, the throughput of the proposed scheme using the sequential approach is analyzed based on both the soft decision fusion rule and hard decision fusion rule. Simulation results indicate that the new approach improves primary user sensing accuracy by 13% over previous approaches, while concurrently increasing detection probability and decreasing false alarm probability. It also improves overall throughput, reduces energy consumption, prolongs expected lifetime, and reduces global error probability compared to the previous approaches under any condition [part of this paper was presented at the EuCAP2018 conference (Md. Sipon Miah et al. 2018)].

Md. Sipon Miah (Co-Author)

Abstract-For the purpose of attaining a high degree of freedom (DOF) for the direction of arrival (DOA) estimations in radar technology, coprime sensor arrays (CSAs) are evaluated in this paper. In addition, the global and local minima of extremely non-linear functions are investigated, aiming to improve DOF. The optimization features of the cuckoo search (CS) algorithm are utilized for DOA estimation of far-field sources in a low signal-to-noise ratio (SNR) environment. The analytical approach of the proposed CSAs, CS and global and local minima in terms of cumulative distribution function (CDF), fitness function and SNR for DOA accuracy are presented. The parameters like root mean square error (RMSE) for frequency distribution, RMSE variability analysis, estimation accuracy, RMSE for CDF, robustness against snapshots and noise and RMSE for Monte Carlo simulation runs are explored for proposed model performance estimation. In conclusion, the proposed DOA estimation in radar technology through CS and CSA achievements are contrasted with existing tools such as particle swarm optimization (PSO).

Data Privacy and Security, Machine Learning, Big Data Analytics, Online Social Networks Analysis, and Blockchain Technology

0

https://scholar.google.com/citations?user=ZO0t9ZcAAAAJ&hl=en

Recently, Personal Data Storage (PDS) has inaugurated a substantial change to the way people can store and control their personal data, by moving from a service-centric to a user-centric model. PDS offers individuals the capability to keep their data in a unique logical repository, that can be connected and exploited by proper analytical tools, or shared with third parties under the control of end users. Up to now, most of the research on PDS has focused on how to enforce user privacy preferences and how to secure data when stored into the PDS. In contrast, in this paper we aim at designing a Privacy-aware Personal Data Storage (P-PDS), that is, a PDS able to automatically take privacy-aware decisions on third parties access requests in accordance with user preferences. The proposed P-PDS is based on preliminary results presented in [1], where it has been demonstrated that semi-supervised learning can be successfully exploited to make a PDS able to automatically decide whether an access request has to be authorized or not. In this paper, we have deeply revised the learning process so as to have a more usable P-PDS, in terms of reduced effort for the training phase, as well as a more conservative approach w.r.t. users privacy, when handling conflicting access requests. We run several experiments on a realistic dataset exploiting a group of 360 evaluators. The obtained results show the effectiveness of the proposed approach.

Studies on friendships in online social networks involving geographic distance have so far relied on the city location provided in users’ profiles. Consequently, most of the research on friendships has provided accuracy at the city level, at best, to designate a user’s location. This study analyses a Twitter dataset because it provides the exact geographic distance between corresponding users. We start by introducing a strong definition of ‘friend’ on Twitter (i.e. a definition of bidirectional friendship), requiring bidirectional communication. Next, we utilize geo-tagged mentions delivered by users to determine their locations, where ‘@username’ is contained anywhere in the body of tweets. To provide analysis results, we first introduce a friend-counting algorithm. From the fact that Twitter users are likely to post consecutive tweets in the static mode, we also introduce a two-stage distance-estimation algorithm. As the first of our main contributions, we verify that the number of friends of a particular Twitter user follows a well-known power-law distribution (i.e. a Zipf’s distribution or a Pareto distribution). Our study also provides the following newly discovered friendship degree related to the issue of space: the number of friends according to distance follows a double power-law (i.e. a double Pareto law) distribution, indicating that the probability of befriending a particular Twitter user is significantly reduced beyond a certain geographic distance between users, termed the separation point. Our analysis provides concrete evidence that Twitter can be a useful platform for assigning a more accurate scalar value to the degree of friendship between two users.

The concept of Personal Data Storage (PDS) has recently emerged as an alternative and innovative way of managing personal data w.r.t. the service-centric one commonly used today. The PDS offers a unique logical repository, allowing individuals to collect, store, and give access to their data to third parties. The research on PDS has so far mainly focused on the enforcement mechanisms, that is, on how user privacy preferences can be enforced. In contrast, the fundamental issue of preference specification has been so far not deeply investigated. In this paper, we do a step in this direction by proposing different learning algorithms that allow a fine-grained learning of the privacy aptitudes of PDS owners. The learned models are then used to answer third party access requests. The extensive experiments we have performed show the effectiveness of the proposed approach.

Personal data storages (PDSs) give individuals the ability to store their personal data in a data unified repository and control release of their data to data consumers. Being able to gather personal data from different data sources (e.g., banks, hospitals), PDSs will play strategic role in individual privacy management. As such, PDS demands for new privacy models for protecting personal data. In this paper, we propose a new technical approach that empowers individuals to better control data in PDS. Particularly, we present a privacy-aware PDS architecture by focusing on two logical data zones based on the categories of personal data. Moreover, we propose a strategy for regulating personal data release that takes in consideration both user preferences and possible risks and benefits of the data release.

In recent decades, human brain tumor detection has become one of the most challenging issues in medical science. In this paper, we propose a model that includes the template-based K means and improved fuzzy C means (TKFCM) algorithm for detecting human brain tumors in a magnetic resonance imaging (MRI) image. In this proposed algorithm, firstly, the template-based K-means algorithm is used to initialize segmentation significantly through the perfect selection of a template, based on gray-level intensity of image; secondly, the updated membership is determined by the distances from cluster centroid to cluster data points using the fuzzy C-means (FCM) algorithm while it contacts its best result, and finally, the improved FCM clustering algorithm is used for detecting tumor position by updating membership function that is obtained based on the different features of tumor image including Contrast, Energy, Dissimilarity, Homogeneity, Entropy, and Correlation. Simulation results show that the proposed algorithm achieves better detection of abnormal and normal tissues in the human brain under small detachment of gray-level intensity. In addition, this algorithm detects human brain tumors within a very short time—in seconds compared to minutes with other algorithms.

Automated social behavior analysis in the mammalian animal has become an increasingly popular and attractive alternative to traditional manual human annotation with the advancement of machine learning and video tracking system for automatic detection. In this work, we study a framework of how different features perform on the different classifiers to analyze automatic mice behavior. We conducted experiments on the Caltech Resident-Intruder Mouse (CRIM13) dataset, which provides two types of features: trajectory features and spatio-temporal features. With this feature, we train AdaBoost and Random Decision Forest (TreeBagger) classifiers to classify different mouse behaviors to show which features perform best on which classifier. The experimental result shows that the trajectory features are more informative and provide better accuracy than the widely used spatio-temporal features, and AdaBoost classifier shows better performance than the TreeBagger on these features.

The university admission tests find the applicant's ability to admit to the desired university. Nowadays, there is a huge competition in the university admission tests. The failure in the admission tests makes an examinee depressed. This paper proposes a method that predicts undergraduate admission in universities. It can help students to improve their preparation to get a chance at their desired university. Many factors are responsible for the failure or success in an admission test. Educational data mining helps us to analyze and extract information from these factors. Here, the authors apply three machine learning algorithms XGBoost, LightGBM, and GBM on a collected dataset to estimate the probability of getting admission to the university after attending or before attending the admission test. They also evaluate and compare the performance levels of these three algorithms based on two different evaluation metrics – accuracy and F1 score. Furthermore, the authors explore the important factors which influence predicting undergraduate admission.

A smart environment consisting of sensors and actuators that network a physical space makes the lives of users extremely convenient. In such environments, service providers collect, store, and process a huge amount of personal information from the users, and provide appropriate services in return. Processing personal information helps service providers dispense personalized services to users. The flow of personal information requires legal documents such as privacy preferences and privacy policies from the users and service providers, respectively. A software agent (to be installed either at the client’s or server’s end) performs the privacy checking (i.e., matches the service provider’s privacy policies with the user’s privacy preferences), and produces a decision based on a “take it or leave it” approach. Often, this check is too restrictive such that the user is denied the services, even if the privacy policies partially satisfy the user preferences. In this work, the privacy checking problem is envisaged as a knapsack problem, and to address it, we derive (i) knapsack privacy checking (KPC), and (ii) knapsack graph-based privacy checking (KPC-G) techniques. The system developed using these techniques finds the services that best match the user’s privacy preferences as well as the service provider’s policies, with a twofold functionality: (a) it recommends the best matching services based on user-supplied tolerance values, or (b) it solves an optimization problem to suggest services after adjusting the user-supplied tolerance values. Our experimental results demonstrate the effectiveness of the proposed model.

Human civilization is experiencing a critical situation that presents itself for a new coronavirus disease 2019 (COVID-19). This virus emerged in late December 2019 in Wuhan city, Hubei, China. The grim fact of COVID-19 is, it is highly contagious in nature, therefore, spreads rapidly all over the world and causes severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Responding to the severity of COVID-19 research community directs the attention to the analysis of COVID-19, to diminish its antagonistic impact towards society. Numerous studies claim that the subcontinent, i.e., Bangladesh, India, and Pakistan, could remain in the worst affected region by the COVID-19. In order to prevent the spread of COVID-19, it is important to predict the trend of COVID-19 beforehand the planning of effective control strategies. Fundamentally, the idea is to dependably estimate the reproduction number to judge the spread rate of COVID-19 in a particular region. Consequently, this paper uses publicly available epidemiological data of Bangladesh, India, and Pakistan to estimate the reproduction numbers. More specifically, we use various models (for example, susceptible infection recovery (SIR), exponential growth (EG), sequential Bayesian (SB), maximum likelihood (ML) and time dependent (TD)) to estimate the reproduction numbers and observe the model fitness in the corresponding data set. Experimental results show that the reproduction numbers produced by these models are greater than 1.2 (approximately) indicates that COVID-19 is gradually spreading in the subcontinent.

Malicious websites are increasing in abundance, which brings serious web security threats to users. As a result, individuals lose their assets, values, information, etc. to unauthorized parties and become victims while visiting such websites. The research community put efforts into developing effective and efficient models for detecting malicious URLs to make available notifications about websites that users access. Perceiving this, numerous methods make use of various types of machine learning (ML) approaches. However, until now, no technique has perfectly detected malicious URLs, as they are susceptible to false positive and false negative decisions in classifying URLs in malicious and non-malicious groups. To improve this issue, this article proposes a new approach based on a machine learning technique. The proposed approach uses multiple classifiers (i.e., ensemble learning) to predict the class probabilities of URLs and then applies a threshold to filter the decisions of multiple classifiers. Next, the decisions are combined concerning its corresponding class probabilities and find the class label with the highest class probability as the final decision on unlabeled URLs. The results show that the proposed method offers better performance in terms of malicious URL detection than other methods.

0

Computer Vision, Image Processing, Deep Learning, Machine Learning

0

https://scholar.google.com/citations?user=B0kyLGkAAAAJ&hl=en

0

MANET, BigData, IoT, Blockchain Communication

0