rss_2.0Cybernetics and Information Technologies FeedSciendo RSS Feed for Cybernetics and Information Technologies and Information Technologies 's Cover Feature Selection Method for Intrusion Detection Systems Based on an Improved Intelligent Water Drop Algorithm<abstract> <title style='display:none'>Abstract</title> <p>A critical task and a competitive research area is to secure networks against attacks. One of the most popular security solutions is Intrusion Detection Systems (IDS). Machine learning has been recently used by researchers to develop high performance IDS. One of the main challenges in developing intelligent IDS is Feature Selection (FS). In this manuscript, a hybrid FS for the IDS network is proposed based on an ensemble filter, and an improved Intelligent Water Drop (IWD) wrapper. The Improved version from IWD algorithm uses local search algorithm as an extra operator to increase the exploiting capability of the basic IWD algorithm. Experimental results on three benchmark datasets “UNSW-NB15”, “NLS-KDD”, and “KDDCUPP99” demonstrate the effectiveness of the proposed model for IDS versus some of the most recent IDS algorithms existing in the literature depending on “F-score”, “accuracy”, “FPR”, “TPR” and “the number of selected features” metrics.</p> </abstract>ARTICLE2022-11-10T00:00:00.000+00:00B-Morpher: Automated Learning of Morphological Language Characteristics for Inflection and Morphological Analysis<abstract> <title style='display:none'>Abstract</title> <p>The automated induction of inflection rules is an important research area for computational linguistics. In this paper, we present a novel morphological rule induction model called B-Morpher that can be used for both inflection analysis and morphological analysis. The core element of the engine is a modified Bayes classifier in which class categories correspond to general string transformation rules. Beside the core classification module, the engine contains a neural network module and verification unit to improve classification accuracy. For the evaluation, beside the large Hungarian dataset the tests include smaller non-Hungarian datasets from the SIGMORPHON shared task pools. Our evaluation shows that the efficiency of B-Morpher is comparable with the best results, and it outperforms the state-of-theart base models for some languages. The proposed system can be characterized by not only high accuracy, but also short training time and small knowledge base size.</p> </abstract>ARTICLE2022-11-10T00:00:00.000+00:00Copy-Move Forgery Detection Using Superpixel Clustering Algorithm and Enhanced GWO Based AlexNet Model<abstract> <title style='display:none'>Abstract</title> <p>In this work a model is introduced to improve forgery detection on the basis of superpixel clustering algorithm and enhanced Grey Wolf Optimizer (GWO) based AlexNet. After collecting the images from MICC-F600, MICC-F2000 and GRIP datasets, patch segmentation is accomplished using a superpixel clustering algorithm. Then, feature extraction is performed on the segmented images to extract deep learning features using an enhanced GWO based AlexNet model for better forgery detection. In the enhanced GWO technique, multi-objective functions are used for selecting the optimal hyper-parameters of AlexNet. Based on the obtained features, the adaptive matching algorithm is used for locating the forged regions in the tampered images. Simulation outcome showed that the proposed model is effective under the conditions: salt &amp; pepper noise, Gaussian noise, rotation, blurring and enhancement. The enhanced GWO based AlexNet model attained maximum detection accuracy of 99.66%, 99.75%, and 98.48% on MICC-F600, MICC-F2000 and GRIP datasets.</p> </abstract>ARTICLE2022-11-10T00:00:00.000+00:00A Robust Feature Construction for Fish Classification Using Grey Wolf Optimizer<abstract> <title style='display:none'>Abstract</title> <p>The low quality of the collected fish image data directly from its habitat affects its feature qualities. Previous studies tended to be more concerned with finding the best method rather than the feature quality. This article proposes a new fish classification workflow using a combination of Contrast-Adaptive Color Correction (NCACC) image enhancement and optimization-based feature construction called Grey Wolf Optimizer (GWO). This approach improves the image feature extraction results to obtain new and more meaningful features. This article compares the GWO-based and other optimization method-based fish classification on the newly generated features. The comparison results show that GWO-based classification had 0.22% lower accuracy than GA-based but 1.13 % higher than PSO. Based on ANOVA tests, the accuracy of GA and GWO were statistically indifferent, and GWO and PSO were statistically different. On the other hand, GWO-based performed 0.61 times faster than GA-based classification and 1.36 minutes faster than the other.</p> </abstract>ARTICLE2022-11-10T00:00:00.000+00:00Modelling and Forecasting of EUR/USD Exchange Rate Using Ensemble Learning Approach<abstract> <title style='display:none'>Abstract</title> <p>The aim of the study is to obtain an accurate result from forecasting the EUR/USD exchange rate. To this end, high-performance machine learning models using CART Ensembles and Bagging method have been developed. Key macroeconomic indicators have been also examined including inflation in Europe and the United States, the index of unemployment in Europe and the United States, and more. Official monthly data in the period from December 1998 to December 2021 have been studied. A careful analysis of the macroeconomic time series has shown that their lagged variables are suitable for model’s predictors. CART Ensembles and Bagging predictive models having been built, explaining up to 98.8% of the data with MAPE of 1%. The degree of influence of the considered macroeconomic indicators on the EUR/USD rate has been established. The models have been used for forecasting one-month-ahead. The proposed approach could find a practical application in professional trading, budgeting and currency risk hedging.</p> </abstract>ARTICLE2022-11-10T00:00:00.000+00:00A Decentralized Medical Network for Maintaining Patient Records Using Blockchain Technology<abstract> <title style='display:none'>Abstract</title> <p>Personal Medical Records (PMR) manage an individual’s medical information in digital form and allow patients to view their medical information and doctors to diagnose diseases. Today’s institution-dependent centralized storage, fails to give trustworthy, secure, reliable, and traceable patient controls. This leads to a serious disadvantage in diagnosing and preventing diseases. The proposed blockchain technique forms a secured network between doctors of the same specialization for gathering opinions on a particular diagnosis by sharing the PMR with consent to provide better care to patients. To finalize the disease prediction, members can approve the diagnosis. The smart contract access control allows doctors to view and access the PMR. The scalability issue is resolved by the Huffman code data compression technique, and security of the PMR is achieved by an advanced encryption standard. The proposed techniques’ requirements, latency time, compression ratio and security analysis have been compared with existing techniques.</p> </abstract>ARTICLE2022-11-10T00:00:00.000+00:00A Model-Free Cognitive Anti-Jamming Strategy Using Adversarial Learning Algorithm<abstract> <title style='display:none'>Abstract</title> <p>Modern networking systems can benefit from Cognitive Radio (CR) because it mitigates spectrum scarcity. CR is prone to jamming attacks due to shared communication medium that results in a drop of spectrum usage. Existing solutions to jamming attacks are frequently based on Q-learning and deep Q-learning networks. Such solutions have a reputation for slow convergence and learning, particularly when states and action spaces are continuous. This paper introduces a unique reinforcement learning driven anti-jamming scheme that uses adversarial learning mechanism to counter hostile jammers. A mathematical model is employed in the formulation of jamming and anti-jamming strategies based on deep deterministic policy gradients to improve their policies against each other. An open-AI gym-oriented customized environment is used to evaluate proposed solution concerning power-factor and signal-to-noise-ratio. The simulation outcome shows that the proposed anti-jamming solution allows the transmitter to learn more about the jammer and devise the optimal countermeasures than conventional algorithms.</p> </abstract>ARTICLE2022-11-10T00:00:00.000+00:00An Approach for Securing JSON Objects through Chaotic Synchronization<abstract> <title style='display:none'>Abstract</title> <p>Nowadays the interoperability of web applications is carried out by the use of data exchange formats such as XML and JavaScript Object Notation (JSON). Due to its simplicity, JSON objects are the most common way for sending information over the HTTP protocol. With the aim of adding a security mechanism to JSON objects, in this work we propose an encryption approach for cipher JSON objects through the use of chaotic synchronization. Synchronization ability between two chaotic systems offers the possibility of securing information between two points. Our approach includes mechanisms for diffusing and confusing JSON objects (plaintext), which yields a proper ciphertext. Our approach can be applied as an alternative to the existing securing JSON approaches such as JSON Web Encryption (JWE).</p> </abstract>ARTICLE2022-11-10T00:00:00.000+00:00A New Attribute-Based Access Control Model for RDBMS<abstract> <title style='display:none'>Abstract</title> <p>One of the challenges in Attribute-Based Access Control (ABAC) implementation is acquiring sufficient metadata against entities and attributes. Intelligent mining and extracting ABAC policies and attributes make ABAC implementation more feasible and cost-effective. This research paper focuses on attribute extraction from an existing enterprise relational database management system – RDBMS. The proposed approach tends to first classify entities according to some aspects of RDBMS systems. By reverse engineering, some metadata elements and ranking values are calculated for each part. Then entities and attributes are assigned a final rank that helps to decide what attribute subset is a candidate to be an optimal input for ABAC implementation. The proposed approach has been tested and implemented against an existing enterprise RDBMS, and the results are then evaluated. The approach enables the choice to trade-off between accuracy and overhead. The results score an accuracy of up to 80% with no overhead or 88% of accuracy with 65% overhead.</p> </abstract>ARTICLE2022-11-10T00:00:00.000+00:00Fuzzy Neutrosophic Soft Set Based Transfer-Q-Learning Scheme for Load Balancing in Uncertain Grid Computing Environments<abstract> <title style='display:none'>Abstract</title> <p>Effective load balancing is tougher in grid computing compared to other conventional distributed computing platforms due to its heterogeneity, autonomy, scalability, and adaptability characteristics, resource selection and distribution mechanisms, and data separation. Hence, it is necessary to identify and handle the uncertainty of the tasks and grid resources before making load balancing decisions. Using two potential forms of Hidden Markov Models (HMM), i.e., Profile Hidden Markov Model (PF_HMM) and Pair Hidden Markov Model (PR_HMM), the uncertainties in the task and system parameters are identified. Load balancing is then carried out using our novel Fuzzy Neutrosophic Soft Set theory (FNSS) based transfer Q-learning with pre-trained knowledge. The transfer Q-learning enabled with FNSS solves large scale load balancing problems efficiently as the models are already trained and do not need pre-training. Our expected value analysis and simulation results confirm that the proposed scheme is 90 percent better than three of the recent load balancing schemes.</p> </abstract>ARTICLE2022-11-10T00:00:00.000+00:00Optimal High Pass FIR Filter Based on Adaptive Systematic Cuckoo Search Algorithm<abstract> <title style='display:none'>Abstract</title> <p>This paper presents the design of a desired linear phase digital Finite Impulse Response (FIR) High Pass (HP) filter based on Adaptive Systematic Cuckoo Search Algorithm (ACSA). The deviation, or error from the desired response, is assessed along with the stop-band and pass-band attenuation of the filter. The Cuckoo Search algorithm (CS) is used to avoid local minima because the error surface is typically non-differentiable, nonlinear, and multimodal. The ACSA is applied to the minimax criterion (L∞-norm) based error fitness function, which offers a better equiripple response for passband and stopband, high stopband attenuation, and rapid convergence for the developed optimal HP FIR filter algorithm. The simulation findings demonstrate that when compared to the Parks McClellan (PM), Particle Swarm Optimization (PSO), CRazy Particle Swarm Optimization (CRPSO), and Cuckoo Search algorithms, the proposed HP FIR filter employing ACSA leads to better solutions.</p> </abstract>ARTICLE2022-11-10T00:00:00.000+00:00One-vs-All Convolutional Neural Networks for Synthetic Aperture Radar Target Recognition<abstract> <title style='display:none'>Abstract</title> <p>Convolutional Neural Networks (CNN) have been widely utilized for Automatic Target Recognition (ATR) in Synthetic Aperture Radar (SAR) images. However, a large number of parameters and a huge training data requirements limit CNN’s use in SAR ATR. While previous works have primarily focused on model compression and structural modification of CNN, this paper employs the One-Vs-All (OVA) technique on CNN to address these issues. OVA-CNN comprises several Binary classifying CNNs (BCNNs) that act as an expert in correctly recognizing a single target. The BCNN that predicts the highest probability for a given target determines the class to which the target belongs. The evaluation of the model using various metrics on the Moving and Stationary Target Acquisition and Recognition (MSTAR) benchmark dataset illustrates that the OVA-CNN has fewer weight parameters and training sample requirements while exhibiting a high recognition rate.</p> </abstract>ARTICLE2022-09-22T00:00:00.000+00:00Toward Programmability of Radio Resource Control Based on O-RAN<abstract> <title style='display:none'>Abstract</title> <p>Open Radio Access Network (O-RAN) is a concept that aims at embedding intelligence at the network edge and at disaggregating of network functionality from the hardware. The paper studies how the O-RAN concept can be used for optimization of radio resource management. The research focuses on adaptive radio resource allocation based on predictions of device activity. For narrowband devices which send sporadically small volumes of data, a feature is defined which enables a device with no activity for a short time to suspend its session and to resume it moving in active state. Dynamic configuration of the inactivity timer based on prediction of device activity may further optimize radio resource allocation. The paper studies an O-RAN use case for dynamic radio resource control and presents the results of emulation of the RESTful interface defined between the O-RAN non-real-time and near real-time functions.</p> </abstract>ARTICLE2022-09-22T00:00:00.000+00:00Investigation of Dense Family of Closure Operations<abstract> <title style='display:none'>Abstract</title> <p>As a basic notion in algebra, closure operations have been successfully applied to many fields of computer science. In this paper we study dense family in the closure operations. In particular, we prove some families to be dense in any closure operation, in which the greatest and smallest dense families, including the collection of the whole closed sets and the minimal generator of the closed sets, are also pointed out. More important, a necessary and sufficient condition for an arbitrary family to be dense is provided in our paper. Then we use these dense families to characterize minimal keys of the closure operation under the viewpoint of transversal hypergraphs and construct an algorithm for determining the minimal keys of a closure operation.</p> </abstract>ARTICLE2022-09-22T00:00:00.000+00:00A New Network Digital Forensics Approach for Internet of Things Environment Based on Binary Owl Optimizer<abstract> <title style='display:none'>Abstract</title> <p>The Internet of Things (IoT) is widespread in our lives these days (e.g., Smart homes, smart cities, etc.). Despite its significant role in providing automatic real-time services to users, these devices are highly vulnerable due to their design simplicity and limitations regarding power, CPU, and memory. Tracing network traffic and investigating its behavior helps in building a digital forensics framework to secure IoT networks. This paper proposes a new Network Digital Forensics approach called (NDF IoT). The proposed approach uses the Owl optimizer for selecting the best subset of features that help in identifying suspicious behavior in such environments. The NDF IoT approach is evaluated using the Bot IoT UNSW dataset in terms of detection rate, false alarms, accuracy, and f-score. The approach being proposed has achieved 100% detection rate and 99.3% f-score and outperforms related works that used the same dataset while reducing the number of features to three features only.</p> </abstract>ARTICLE2022-09-22T00:00:00.000+00:00Information Systems Reliability in Traditional Entropy and Novel Hierarchy<abstract> <title style='display:none'>Abstract</title> <p>The continuous progress of computing technologies increases the need for improved methods and tools for assessing the performance of information systems in terms of reliability, conformance, and quality of service. This paper presents an extension of Information Theory by introducing a novel hierarchy concept as a complement to the traditional entropy approach. The methodology adjustments are applied to a simulative numerical example for assessing the reliability of systems with different complexity and performance behavior.</p> </abstract>ARTICLE2022-09-22T00:00:00.000+00:00Mathematical Modelling of Malware Intrusion in Computer Networks<abstract> <title style='display:none'>Abstract</title> <p>Malware attacks cause great harms in the contemporary information systems and that requires analysis of computer networks reaction in case of malware impact. The focus of the present study is on the analysis of the computer network’s states and reactions in case of malware attacks defined by the susceptibility, exposition, infection and recoverability of computer nodes. Two scenarios are considered – equilibrium without secure software and not equilibrium with secure software in the computer network. The behavior of the computer network under a malware attack is described by a system of nonhomogeneous differential equations. The system of the nonhomogeneous differential equations is solved, and analytical expressions are derived to analyze network characteristics in case of susceptibility, exposition, infection and recoverability of computer nodes during malware attack. The analytical expressions derived are illustrated with results of numerical experiments. The conception developed in this work can be applied to control, prevent and protect computer networks from malware intrusions.</p> </abstract>ARTICLE2022-09-22T00:00:00.000+00:00A Color-Texture-Based Deep Neural Network Technique to Detect Face Spoofing Attacks<abstract> <title style='display:none'>Abstract</title> <p>Given the face spoofing attack, adequate protection of human identity through face has become a significant challenge globally. Face spoofing is an act of presenting a recaptured frame before the verification device to gain illegal access on behalf of a legitimate person with or without their concern. Several methods have been proposed to detect face spoofing attacks over the last decade. However, these methods only consider the luminance information, reflecting poor discrimination of spoofed face from the genuine face. This article proposes a practical approach combining Local Binary Patterns (LBP) and convolutional neural network-based transfer learning models to extract low-level and high-level features. This paper analyzes three color spaces (i.e., RGB, HSV, and YCrCb) to understand the impact of the color distribution on real and spoofed faces for the NUAA benchmark dataset. In-depth analysis of experimental results and comparison with other existing approaches show the superiority and effectiveness of our proposed models.</p> </abstract>ARTICLE2022-09-22T00:00:00.000+00:00Serverless High-Performance Computing over Cloud<abstract> <title style='display:none'>Abstract</title> <p>HPC clouds may provide fast access to fully configurable and dynamically scalable virtualized HPC clusters to address the complex and challenging computation and storage-intensive requirements. The complex environmental, software, and hardware requirements and dependencies on such systems make it challenging to carry out our large-scale simulations, prediction systems, and other data and compute-intensive workloads over the cloud. The article aims to present an architecture that enables HPC workloads to be serverless over the cloud (Shoc), one of the most critical cloud capabilities for HPC workloads. On one hand, Shoc utilizes the abstraction power of container technologies like Singularity and Docker, combined with the scheduling and resource management capabilities of Kubernetes. On the other hand, Shoc allows running any CPU-intensive and data-intensive workloads in the cloud without needing to manage HPC infrastructure, complex software, and hardware environment deployments.</p> </abstract>ARTICLE2022-09-22T00:00:00.000+00:00Hardware Response and Performance Analysis of Multicore Computing Systems for Deep Learning Algorithms<abstract> <title style='display:none'>Abstract</title> <p>With the advancement in technological world, the technologies like Artificial Intelligence (AI), Machine Learning (ML), and Deep Learning (DL) are gaining more popularity in many applications of computer vision like object classification, object detection, Human detection, etc., ML and DL approaches are highly compute-intensive and require advanced computational resources for implementation. Multicore CPUs and GPUs with a large number of dedicated processor cores are typically the more prevailing and effective solutions for the high computational need. In this manuscript, we have come up with an analysis of how these multicore hardware technologies respond to DL algorithms. A Convolutional Neural Network (CNN) model have been trained for three different classification problems using three different datasets. All these experimentations have been performed on three different computational resources, i.e., Raspberry Pi, Nvidia Jetson Nano Board, &amp; desktop computer. Results are derived for performance analysis in terms of classification accuracy and hardware response for each hardware configuration.</p> </abstract>ARTICLE2022-09-22T00:00:00.000+00:00en-us-1