Malware detection using Machine Learning techniques has gained popularity due to their high accuracy. However, ML models are susceptible to Adversarial Examples, specifically crafted samples intended to deceive the detectors. This paper presents a novel method for generating evasive AEs by augmenting existing malware with a new section at the end of the PE file, populated with binary data using memetic algorithms. Our method hybridizes global search and local search techniques to achieve optimized results. The Malconv Model, a well-known state-of-the-art deep learning model designed explicitly for detecting malicious PE files, was used to assess the evasion rates. Out of 100 tested samples, 98 successfully evaded the MalConv model. Additionally, we investigated the simultaneous evasion of multiple detectors, observing evasion rates of 35% and 44% against KNN and Decision Tree machine learning detectors, respectively. Furthermore, evasion rates of 26% and 10% were achieved against Kaspersky and ESET commercial detectors. In order to prove the efficiency of our memetic algorithm in generating evasive adversarial examples, we compared it to the most used evolutionary-based attack: the genetic algorithm. Our method demonstrated significantly superior performance while utilizing fewer generations and a smaller population size.

Data packets in Wireless Mesh Networks (WMNs) are routed across several nodes in a multi-hop fashion. The Quality of Service (QoS), seamless connectivity, reliability, and scalability of Wireless Mesh Networks are all significantly impacted by routing approaches. Routing protocols should enforce the fair utilization of resources i.e. bandwidth or channel among network nodes irrespective of their spatial location from the Gateway. The two-hop or multi-hop nodes in wireless mesh networks experience resource starvation due to the functioning of the MAC protocol and TCP/TP networking protocol. The Starvation issue has a significant impact on the QoS requirements of wireless mesh networks. It is known that using appropriate scheduling techniques in network planning substantially minimizes starvation. To reduce the starving of resources to the multi-hop network nodes, novel optimized routing algorithms have been proposed and implemented in this work. To address the starvation, a GA-based cross-layer optimized scheduling method that operates at the MAC and Network layers is implemented. A hybrid approach that combines the features of the Genetic Algorithm (GA) and Gravitational Search Algorithm (GSA) is also implemented to solve the local minimum problem in GA. Results show that the suggested optimization methods greatly improve the fairness performance of wireless mesh networks.

This research investigates the performance of various one-dimensional cellular automata rules, namely Rule 30, Rule 150, Rule 184, Rule 105 and Rule 110, for image encryption. The proposed algorithm technique combines these CA rules to generate pseudo-random sequences for image scrambling. The effectiveness of the proposed method is evaluated using various performance metrics, including NPCR, correlation coefficient and information entropy. The results demonstrate which rule provides the best encryption performance, achieving high levels of security and resistance to statistical attacks. However, the computational complexity of the proposed method is relatively high, which may limit its practicality for real-time image encryption applications.

The rise of cyberattacks has led to an increase in the creation of fake websites by attackers, who use these sites for advertising products, transmit malware, or steal valuable login credentials. Phishing, the act of soliciting sensitive information from users by masquerading as a trustworthy entity, is a common technique used by attackers to achieve their goals. Spoofed websites and email spoofing are often used in phishing attacks, with spoofed emails redirecting users to phishing websites in order to trick them into revealing their personal information. Traditional solutions for detecting phishing websites rely on signature-based approaches that are not effective in detecting newly created spoofed websites. To address this challenge, researchers have been exploring machine-learning methods for detecting phishing websites. In this paper, we suggest a new approach that combines the use of blacklists and machine learning techniques such that a variety of powerful features, including domain-based features, abnormal features, and abnormal features based on URLs, HTML, and JavaScript, to rank web pages and improve classification accuracy. Our experimental results show that using the proposed approach, the random forest classifier offers the best accuracy of 93%, with FPR and FNR as 0.12 and 0.02, with a Precision of 90%, Recall of 97% an F1 Score of 93%, and MCC of 0.85.

The critical challenge with the continuously increasing number of Internet users is copying and duplication, which has caused content integrity and protection. To manage and secure the signals from unauthorized consumers of digital content, we require certain procedures. Digital watermarking scheme on vowel-based approach can address these problems. Thus, we can provide a robust and secure method that solves the issues of copyright, illegal intentional or unintentional modification. In this paper, we have proposed vowel-based speech watermarking techniques using the FFT method with the help of the Min algorithm. We observe that the proposed FFT-based watermarking scheme provides better results in comparison to the existing methods.

Pay-as-you-go models are used to grant users access to cloud services. While using the cloud, an imbalance workload on data centre resources degrades quality of service metrics like makespan, storage, high failure rate, and energy consumption. Hence proposed the heuristic based hybrid GA to enhance the QoS with resource allocation in cloud computing. The population is first initialized using the Binary encoding sorts the tasks according to priority. After that, the Best Fit algorithm compares the Best Fit with iterations of each fitness value depending on the computation time to shorten the make span. Heuristic crossover approach and mutation are then used to update the probability of the existing population with the new population lowers the failure rate by using the fitness value. Therefore, the proposed heuristic-based hybrid GA technique balanced the load and allocate the resources effectively to improve QoS performances. The outcome reveals that the proposed method of QoS performances attained less makespan, energy consumption, failure rate and execution time with effectively allocated the resources of 1% to 39% when compared to the previous methods in cloud computing.

An intrusion detection system (IDS) is either a part of a software or hardware environment that monitors data and analyses it to identify any attacks made against a system or a network. Traditional IDS approaches make the system more complicated and less efficient, because the analytical properties process is difficult and time-consuming. This is because the procedure is complex. Therefore, this research work focuses on a network intrusion detection and classification (NIDCS) system using a modified convolutional neural network (MCNN) with recursive feature elimination (RFE). Initially, the dataset is balanced with the help of the local outlier factor (LOF), which finds anomalies and outliers by comparing the amount of deviation that a single data point has with the amount of deviation that its neighbors have. Then, a feature extraction selection approach named RFE is applied to eliminate the weakest features until the desired number of features is achieved. Finally, the optimal features are trained with the MCNN classifier, which classifies intrusions like probe, denial-of-service (DoS), remote-to-user (R2U), user-to-root (U2R), and identifies normal data. The proposed NIDCS system resulted in higher performance with 99.3% accuracy and a 3.02 false alarm rate (FAR) as equated to state-of-the-art NIDCS approaches such as deep neural networks (DNN), ResNet, and gravitational search algorithms (GSA).

The project envisages the creation of a complex system that integrates advanced technologies of machine learning and natural language processing for media content analysis. The main goal is to provide means for quick and accurate verification of information, reduce the impact of disinformation campaigns and increase media literacy of the population. Research tasks included the development of algorithms for the analysis of textual information, the creation of a database of fakes, and the development of an interface for convenient access to analytical tools. The object of the study was the process of spreading information in the media space, and the subject was methods and means for identifying disinformation. The scientific novelty of the project consists of the development of algorithms adapted to the peculiarities of the Ukrainian language, which allows for more effective work with local content and ensures higher accuracy in identifying fake news. Also, the significance of the project is enhanced by its practical value, as the developed tools can be used by government structures, media organizations, educational institutions and the public to increase the level of information security. Thus, the development of this project is of great importance for increasing Ukraine's resilience to information threats and forming an open, transparent information society.