The paper describes the search for a solution to the image near-duplicate detection problem. We assume that there are only two images to compare and classify whether they are near-duplicates. There are some traditional methods to match pair of images, and the evaluation of the most famous of them in terms of the problem is performed in this research. The effective thresholds to separate near-duplicate classes are found during experimental modeling using the INRIA Holidays dataset. The sequence of methods is proposed to make the joint decision better in terms of accuracy. It is shown also that the accuracy of binary classification of the proposed approach for the combination of the histogram comparison and ORB descriptors matching is about 85% for both near-duplicate and not near-duplicate pairs of images. This is compared to the existing methods, and it is shown, that the accuracy of more powerful methods, based on deep learning, is better, but the speed of the proposed method is higher.

Continuous Hopfield network is a recurring network that has shown its ability to solve important optimization problems. Continuous Hopfield network dynamic system is characterized by a differential equation. This equation is difficult to solve, especially for large problems. This led researchers to discretize the differential equation using Euler’s method. However, this method generally does not converge to a good solution because it is sensitive to the step size decision and initial conditions. In this work, we discretize the dynamic system of continuous Hopfield network by a new method of Runge-Kutta. This method is strong in terms of stability and performance in order to converge to a better solution. This new method introduces two phases for better network stability. The first phase targets to solve the dynamic equation by the Euler method, while the second phase allows refining the solution found in the first phase. Experimental results on benchmarks show that the proposed approach can effectively improve Hopfield neural network performance.

In this paper, a method of decreasing the redundancy of information flow by using recurrent properties of Galois code sequences is proposed. For this purpose, the service information is compiled and the priority compression is identified. The method is based on applying one of the adaptive algorithms (prediction first-order, interpolation zero-order, interpolation first-order) by comparing the efficiency of its use when applied to the selected fragments of a signal. It is shown that the developed method is effective for the quick-change signals when the structure and behavior of a signal change drastically. The efficiency of redundancy decreasing at the different sampling rate and the number of the significant samples is evaluated. This makes it possible to establish the limits of the positive effect for redundancy of information flows for the existing and developed methods. Experimental research is carried out for various permissible deviations with obtaining the number of the significant readings. A comparison of the obtained data with results of applying the existing methods in deep pumping installations proved that the proposed method is in 1.3 times more effective than existing ones.

Type 2 diabetes is a well-known lifelong condition disease that reduces the human body’s ability to produce insulin. This causes high blood sugar levels, which leads to different complications, including stroke, eye, cardiovascular, kidney, and nerve damage. Although diabetes has attained the attention of huge research, the classification performance of such medical problems utilizing techniques of machine learning is quite low, primarily due to the class imbalance and the presence of missing values in data. In this work, we proposed a model using binary Grey wolf optimization (GWO) and a Decision tree. The proposed model is composed of preprocessing, feature selection, and classification. In preprocessing, that is responsible for minority class oversampling and handling missing values. In the second step, binary GWO are used to select the most significant features. In the third step, the proposed model is trained using the Decision tree algorithm. The model achieved an accuracy of 83.11% when it was applied on the Pima Indian`s dataset.

Software Defined Network (SDN) is the new era of networking technology based on a centralized controller that separates the switch hardware from its operating software. The most important challenge is the security of SDN and the most prominent attack is the Distributed Denial of Service (DDoS) attack. Some of the research work done so far detects DDoS attacks using a threshold, which is usually assumed without proper scientific reason and hence may not be always accurate. The mitigation techniques used by some researchers block the host from sending the network traffic beyond a threshold, by installing drop rules in the flow table of the switch connected to that host. Doing so will not only block the attack traffic but also the genuine ones from other applications of that host. In this paper, we propose a model that calculates the threshold limit for the type of applications sending data to a particular switch, in real-time using a machine learning (ML) model, and determines whether that application traffic is DDoS traffic. After the detection, only application type sending DDoS traffic is blocked while other genuine applications are allowed to send the network traffic without any interruption. The use of a dynamic threshold, based on the current network traffic, will help in detecting DDoS efficiently.

The subject of the study in this article is data transmission processes of the video information resource in the information communication systems of the air segment under the conditions of errors in the data transmission channel. The purpose of the article is the development of the method of compression coding in order to ensure an increase in the level of reliability of video information resources under the conditions of errors in communication channels. The following tasks are identified: to develop a method of compression coding using structural decomposition of statistical space; analyze the effectiveness of the developed method from the standpoint of ensuring the required level of reliability. The following results are obtained: the developed method of encoding video information allows increasing the level of reliability in the conditions of the transmission of video information resources in the information communication systems of the air segment due to the localization of the action of errors.

Dynamic metrics capture the run time features of object-oriented languages, i.e., runtime polymorphism, dynamic binding, etc., that are not covered by static metrics. Therefore, in this paper, we derived a new approach to measuring the software reusability of a design pattern based on dynamic metrics. To achieve this, the authors proposed a model based on five parameters, i.e., polymorphism, inheritance, number of children, coupling, and complexity, to measure the reusability factor by using various soft computing techniques, i.e., Fuzzy, Neural Network, and Neuro-Fuzzy. Further, we also compared the proposed model with four existing machine learning algorithms. Lastly, we found that the proposed model using the neuro-fuzzy technique is trained well and predicts well with MAE (Mean absolute error) 0.003 and RMSE (Root mean square error) 0.009 based on dynamic metrics. Hence, it is concluded that dynamic metrics are a better predictor of the reusability factor than static metrics.

User authentication in the online environment is promoting a hugely challenging issue. This has contributed to the realization of a user authentication where the exams can be performed over the Internet at any time and from any place and by using any digital device. Consequently, further investigations are required to focus on improving user authentication methods to enhance online security mechanisms, especially in the field of e-exams. This research proposes a new user authentication technique based on the user interface (UI). The novel idea is created based on the design preferences of candidates who are taking the e-exams. Several design features are used to design a special user interface for e-exams, for example, the font attributes, back colour, number of questions per page, group categories for questions based on difficulties, and timer setting. The introduced technique can be used to support the user authentication process in the e-exams environment. Furthermore, the proposed technique provides the ability to login to the e-exam without the need to remember the login information, but to select what the student prefers according to his/her personal information. Based on the literature review, a primary evaluation claiming that the students have differences in their preferences and that each user has stable design preferences within different sessions is revealed. In regard to these facts, they become the resource and essence of this research. The security performance of the proposed method is evaluated. The results of the experiments show a false positive (FP) rate of 0.416% and a false negative (FN) rate of 0%.

One of the widely used methods to accelerate a numerical solver is implementation of multithreading. The problem of thread allocation on-demand at runtime is latency, caused by periodical instantiation of threads. The article is devoted to parallelization of solver for 3D mathematical model of ore sintering, based on software threads reusing them during computation. Computational domain is equally shared among available threads. Each thread writes only to own data partition. A looped barrier is proposed for guaranteed synchronization of all threads after iteration. The method allows scaling performance without recompilation of the solver by using similar CPU with more cores. Measurement of solver performance with 220 nodes using different thread count confirms scalability around 95% for double and single precision arithmetics. Presented pictures of perspective view with three slices of temperature field show influence of heat loss from pallets walls. A cross section of temperature field in layer after 16 minutes of sintering is calculated with appearance of two high-temperature regions inside. Comparison of temperature field with literature data gives good correspondence. The computer model takes into account important chemical reactions, such as, coke burning, carbonate dissolution, water vaporization, as well as mass-heat transfer inside the sinter layer and can be used in metallurgical plants to increase effectiveness of sintering.

Worldwide, Internet of Things (IoT) devices will surpass a range of five billion by 2025 and developed countries will extend to advance by supplying almost two-thirds of such connections. With existing infrastructure, allocating bandwidth to billions of IoT devices is going to be cumbersome. This paper addresses the problem of Dynamic bandwidth allocation in IoT devices. We enhanced the dynamic bandwidth allocation algorithms to support QoS in different bandwidth ranges. Our Proposed innovative Machine learning-based Intelligent Dynamic Bandwidth Allocation (IDBA) algorithm allocates the bandwidth effectively between IoT devices based on utilization patterns observed through machine learning methods. Moreover, we showed that an IDBA algorithm results in supporting quality of service in terms of ensuring uninterrupted bandwidth to critical IoT application where bandwidth tolerance is zero percent, along with that IDBA increasing the network throughput correlated to other dynamic bandwidth allocation algorithms. We demonstrate simulations in different applications. The results show that IDBA achieves better throughput even in low bandwidth range.