Phishing leverages people’s tendency to share personal information online. Phishing attacks often begin with an email and can be used for a variety of purposes. The cybercriminal will employ social engineering techniques to get the target to click on the link in the phishing email, which will take them to the infected website. These attacks become more complex as hackers personalize their fraud and provide convincing messages. Phishing with a malicious URL is an advanced kind of cybercrime. It might be challenging even for cautious users to spot phishing URLs. The researchers displayed different techniques to address this challenge. Machine learning models improve detection by using URLs, web page content and external features. This article presents the findings of an experimental study that attempted to enhance the performance of machine learning models to obtain improved accuracy for the two phishing datasets that are used the most commonly. Three distinct types of tuning factors are utilized, including data balancing, hyper-parameter optimization and feature selection. The experiment utilizes the eight most prevalent machine learning methods and two distinct datasets obtained from online sources, such as the UCI repository and the Mendeley repository. The result demonstrates that data balance improves accuracy marginally, whereas hyperparameter adjustment and feature selection improve accuracy significantly. The performance of machine learning algorithms is improved by combining all fine-tuned factors, outperforming existing research works. The result shows that tuning factors enhance the efficiency of machine learning algorithms. For Dataset-1, Random Forest (RF) and Gradient Boosting (XGB) achieve accuracy rates of 97.44% and 97.47%, respectively. Gradient Boosting (GB) and Extreme Gradient Boosting (XGB) achieve accuracy values of 98.27% and 98.21%, respectively, for Dataset-2.

The only part of an antigen (a protein molecule found on the surface of a pathogen) that is composed of epitopes specific to T and B cells is recognized by the human immune system (HIS). Identification of epitopes is considered critical for designing an epitope-based peptide vaccine (EBPV). Although there are a number of vaccine types, EBPVs have received less attention thus far. It is important to mention that EBPVs have a great deal of untapped potential for boosting vaccination safety—they are less expensive and take a short time to produce. Thus, in order to quickly contain global pandemics such as the ongoing outbreak of coronavirus disease 2019 (COVID-19) caused by the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), as well as epidemics and endemics, EBPVs are considered promising vaccine types. The high mutation rate of SARS-CoV-2 has posed a great challenge to public health worldwide because either the composition of existing vaccines has to be changed or a new vaccine has to be developed to protect against its different variants. In such scenarios, time being the critical factor, EBPVs can be a promising alternative. To design an effective and viable EBPV against different strains of a pathogen, it is important to identify the putative T- and B-cell epitopes. Using the wet-lab experimental approach to identify these epitopes is time-consuming and costly because the experimental screening of a vast number of potential epitope candidates is required. Fortunately, various available machine learning (ML)-based prediction methods have reduced the burden related to the epitope mapping process by decreasing the potential epitope candidate list for experimental trials. Moreover, these methods are also cost-effective, scalable, and fast. This paper presents a systematic review of various state-of-the-art and relevant ML-based methods and tools for predicting T- and B-cell epitopes. Special emphasis is placed on highlighting and analyzing various models for predicting epitopes of SARS-CoV-2, the causative agent of COVID-19. Based on the various methods and tools discussed, future research directions for epitope prediction are presented.

The sixth generation (6 G) communication standard is expected to include support for very-high data rates (over 100 Gbit/s) and device electronics will require processors with on-chip communications able to support such high bandwidths. Although the terahertz band possesses ample bandwidth, conventional THz waveguides suffer from high bending losses and are sensitive to process defects. The recent revelation of the topological valley photonic crystal (VPC), which exhibits near zero-loss bends, zero back-scattering and zero junction-area, holds much promise for future high speed inter-device communications. Low dispersion in the photonic bandgap region as the number of bends increase is demonstrated through simulation and experiment of the transmission and group delay characteristics. Through comprehensive communications experiments we demonstrate online results below the forward error correction level including an 108-Gbit/s bit rate using multi-level modulation for a 10 mm straight VPC waveguide and a 62.5-Gbit/s bit-rate for a ten sharp bended structure.

Consumer expectations and demands for quality of service (QoS) from network service providers have risen as a result of the proliferation of devices, applications, and services. An exceptional study is being conducted by network design and optimization experts. But despite this, the constantly changing network environment continues to provide new issues that today’s networks must be dealt with effectively. Increased capacity and coverage are achieved by joining existing networks. Mobility management, according to the researchers, is now being investigated in order to make the previous paradigm more flexible, user‐centered, and service‐centric. Additionally, 5G networks provide higher availability, extremely high capacity, increased stability, and improved connection, in addition to quicker speeds and less latency. In addition to being able to fulfil stringent application requirements, the network infrastructure must be more dynamic and adaptive than ever before. Network slicing may be able to meet the present stringent application requirements for network design, if done correctly. The current study makes use of sophisticated fuzzy logic to create algorithms for mobility and traffic management that are as flexible as possible while yet maintaining high performance. Ultimately, the purpose of this research is to improve the quality of service provided by current mobility management systems while also optimizing the use of available network resources. Building SDN (Software‐Defined Networking) and NFV (Network Function Virtualization) technologies is essential. Network slicing is an architectural framework for 5G networks that is intended to accommodate a variety of different networks. In order to fully meet the needs of various use cases on the network, network slicing is becoming more important due to the increasing demand for data rates, bandwidth capacity, and low latency.

The privacy preservation has received considerable attention from organizations as the growing population is apprehensive regarding personal data being preserved.There are several privacy and security issues in existing systems, such as identity and location disclosure, availability, and authenticity. Smart parking systems utilize third-party parking recommender systems to offer customized parking space recommendations to its users based on user’s past parking experience. However, indiscriminately sharing a user’s data with a third party recommendation system may violate their personal information because their activity and node mobility can be deduced from their previous paring experience. Another problem with existing solutions is that most distributed systems need a third party to anonymize user data to preserve the user’s privacy. Therefore, this article provides three solutions that address the problems mentioned above; at first, based on elliptic curve cryptography (ECC), we proposed the mutual authentication mechanism using HMAC to provide anonymity and integrity during communication; secondly, given the risks to security and privacy posed by untrustworthy third parties, we used local differential privacy, which uses the Laplace distribution technique to add noise randomly and eliminates any necessity for a third party for data perturbation. And thirdly, in addition to LDP , we utilize the IOTA distributed ledger technology to provide a new level of security that ensures immutability, scalability, and quantum secrecy and decentralized the system. Our experiments demonstrate that, in addition to preserving the driver’s privacy and security, our proposed model has low storage overheads, computation, and communication costs.

Ransomware is a special malware designed to extort money in return for unlocking the device and personal data files. Smartphone users store their personal as well as official data on these devices. Ransomware attackers found it bewitching for their financial benefits. The financial losses due to ransomware attacks are increasing rapidly. Recent studies witness that out of 87% reported cyber-attacks, 41% are due to ransomware attacks. The inability of application-signature-based solutions to detect unknown malware has inspired many researchers to build automated classification models using machine learning algorithms. Advanced malware is capable of delaying malicious actions on sensing the emulated environment and hence posing a challenge to dynamic monitoring of applications also. Existing hybrid approaches utilize a variety of features combination for detection and analysis. The rapidly changing nature and distribution strategies are possible reasons behind the deteriorated performance of primitive ransomware detection techniques. The limitations of existing studies include ambiguity in selecting the features set. Increasing the feature set may lead to freedom of adept attackers against learning algorithms. In this work, we intend to propose a hybrid approach to identify and mitigate Android ransomware. This study employs a novel dominant feature selection algorithm to extract the dominant feature set. The experimental results show that our proposed model can differentiate between clean and ransomware with improved precision. Our proposed hybrid solution confirms an accuracy of 99.85% with zero false positives while considering 60 prominent features. Further, it also justifies the feature selection algorithm used. The comparison of the proposed method with the existing frameworks indicates its better performance.