Main Modules Research Articles

FMDADM: A Multi-Layer DDoS Attack Detection and Mitigation Framework Using Machine Learning for Stateful SDN-Based IoT Networks

The absence of standards and the diverse nature of the Internet of Things (IoT) have made security and privacy concerns more acute. Attacks such as distributed denial of service (DDoS) are becoming increasingly widespread in IoT, and the need for ways to stop them is growing. The use of newly formed Software-Defined Networking (SDN) significantly lowers the computational burden on IoT network nodes and makes it possible to perform more security measurements. This paper proposes an SDN-based, four-module DDoS attack detection and mitigation framework for IoT networks called FMDADM. The proposed FMDADM framework comprises four main modules and five-tier architecture. The first module implements an early detection process based on the average drop rate (ADR) principle using a 32-packet window size. The second module uses a novel double-check mapping function (DCMF), that aids in earlier attack detection at the data plane level. The third module is an ML-based detection application comprising four phases: data preprocessing, feature extraction, training and testing, and classification. This module detects DDoS attacks using only seven features: two selected and five newly computed features. The last module introduces an attack mitigation process. We applied the proposed framework to three test cases: one single-node attack test case and two multi-node attack test cases, all with real IoT traffic generated and deployed in Mininet-IoT. The proposed FMDADM framework efficiently detects DDoS attacks at high and low rates, can discriminate between attack traffic and flash crowds, and protects both local and remote IoT nodes by preventing infection from propagating to the ISP level. The FMDADM outperformed most existing cutting-edge approaches across ten different evaluation criteria. According to the experimental results, FMDADM achieved the following accuracy, precision, F-measure, recall, specificity, negative predictive value, false positive rate, false detection rate, false negative rate, and average detection time benchmarks:- 99.79%, 99.43%, 99.77%, 99.79%, 99.95%, 00.21%, 00.91%, 00.23%, and 2.64 μs, respectively.

Enhanced Attention-Based Encoder-Decoder Framework for Text Recognition

Recognizing irregular text in natural images is a challenging task in computer vision. The existing approaches still face difficulties in recognizing irregular text because of its diverse shapes. In this paper, we propose a simple yet powerful irregular text recognition framework based on an encoder-decoder architecture. The proposed framework is divided into four main modules. Firstly, in the image transformation module, a Thin Plate Spline (TPS) transformation is employed to transform the irregular text image into a readable text image. Secondly, we propose a novel Spatial Attention Module (SAM) to compel the model to concentrate on text regions and obtain enriched feature maps. Thirdly, a deep bi-directional long short-term memory (Bi-LSTM) network is used to make a contextual feature map out of a visual feature map generated from a Convolutional Neural Network (CNN). Finally, we propose a Dual Step Attention Mechanism (DSAM) integrated with the Connectionist Temporal Classification (CTC) - Attention decoder to re-weights visual features and focus on the intra-sequence relationships to generate a more accurate character sequence. The effectiveness of our proposed framework is verified through extensive experiments on various benchmarks datasets, such as SVT, ICDAR, CUTE80, and IIIT5k. The performance of the proposed text recognition framework is analyzed with the accuracy metric. Demonstrate that our proposed method outperforms the existing approaches on both regular and irregular text. Additionally, the robustness of our approach is evaluated using the grocery datasets, such as GroZi-120, WebMarket, SKU-110K, and Freiburg Groceries datasets that contain complex text images. Still, our framework produces superior performance on grocery datasets.

A Machine Learning-Based Framework for Fast Prediction of Wide-Area Remedial Control Actions in Interconnected Power Systems

This paper presents a novel real-time machinelearning-based framework for remedial control action (RCA) prediction to prevent transient instability in interconnected power systems. Due to the fast dynamics of rotor angle oscillations and considering communication latencies, there is limited time to compute RCA, making RCA calculation impractical in events quickly evolving into transient instability. The proposed algorithm predicts RCA based on pre-fault and post-fault voltage values of generator buses. To cover credible scenarios, reduce prediction complexities, and increase accuracy, a micro model strategy is employed in which independent models are built for each transmission line of the system. The proposed framework consists of three main modules: stability prediction, coherency prediction, and RCA prediction. In the coherency prediction module, a time-varying algorithm is developed that determines the optimal number of generator groups and prevents unnecessarily overestimated RCAs. For each of the considered scenarios and obtained coherency patterns, a mixed-integer linear programming (MILP) model is utilized to extract the islanding and load shedding patterns considering transient stability constraints. The effectiveness of the proposed approach is demonstrated on the IEEE 39-bus system and the 74-bus Nordic test system, followed by a discussion of results.

Building Occupancy Detection and Localization Using CCTV Camera and Deep Learning

Occupancy information plays a key role in analyzing and improving building energy performance. The advances of Internet of Things (IoT) technologies have engendered a shift in measuring building occupancy with IoT sensors, in which cameras in closed-circuit television (CCTV) systems can provide richer measurements. However, existing camera-based occupancy detection approaches cannot function well when scanning videos with a number of occupants and determining occupants’ locations. This article aims to develop a novel deep-learning-based approach for better building occupancy detection based on CCTV cameras. To do so, this research proposes a deep-learning model to detect the number of occupants and determine their locations in videos. This model consists of two main modules, namely, feature extraction and three-stage occupancy detection. The first module presents a deep convolutional neural network to perform residual and multibranch convolutional calculation to extract shallow and semantic features, and constructs feature pyramids through a bidirectional feature network. The second module performs a three-stage detection procedure with three sequential and homogeneous detectors which have increasing Intersection over Union (IoU) thresholds. Empirical experiments evaluate the detection performance of the approach with CCTV videos from a university building. Experimental results show that the approach achieves the superior detection performance when compared with baseline models.

УТВОРЕННЯ ОБ’ЄМНОЇ ФОРМИ ПАКОВАНЬ ТУНЕЛЬНИМ МЕТОДОМ: ОБЛАДНАННЯ І ШЛЯХИ УДОСКОНАЛЕННЯ

he article presents the results of scientific and methodical research in the field of printing and packaging engineering.Speedy production of cardboard packaging, its packing with products, transporting and preservation is a significant factor in successful logistics operations.One of the most widespread is the method, according to which a flat cardboard blank is pushed by a punch into a matrix ("tunnel"), where its flaps are folded into a three-dimensional structure of a tray or box.There is a group of popular designs of trays and boxes according to the FEFCO catalog, which are manufactured by the method.The analysis of publications revealed the existence of both studies related to the phenomena and equipment for folding sheet materials in printing industry, and aimed at improving packaging equipment. The combination of these two directions into a single study is hindered by the currently missing description of the structure and classification of the equipment and its components. The purpose of the article is to analyze and systematize the features of the equipment for volumetric forming of a structure of cardboard containers (boxes and trays), as well as its modules to determine ways of their improvement.A review of the equipment of well-known global manufacturers revealed common features of their modular construction. The mainmechanisms of the cardboard container forming machine are defined, their functions are described, and they are classified according to the features that have the greatest impact on productivity, product quality, and energy saving. A com-parison of application in production was made for the two most widespread schemes of the equipment.Ways to improve the considered equipment and drives of its main mechanisms were outlined.The work describes the stages of the process of volumetric forming of cardboard packings by the tunnel method. The peculiarities of the use of existing devices for the volumetric forming of containers have been revealed. For the first time,the types of equipment for the implementation of the process and its main modules were classified. The presented materials can be used for the rational design of means for the production of cardboard packaging.

Computer Forensics Framework for Efficient and Lawful Privacy-Preserved Investigation

Privacy preservation (PP) in Digital forensics (DF) is a conflicted and non-trivial issue. Existing solutions use the searchable encryption concept and, as a result, are not efficient and support only a keyword search. Moreover, the collected forensic data cannot be analyzed using existing well-known digital tools. This research paper first investigates the lawful requirements for PP in DF based on the organization for economic co-operation and development OECB) privacy guidelines. To have an efficient investigation process and meet the increased volume of data, the presented framework is designed based on the selective imaging concept and advanced encryption standard (AES). The proposed framework has two main modules, namely Selective Imaging Module (SIM) and Selective Analysis Module (SAM). The SIM and SAM modules are implemented based on advanced forensic format 4 (AFF4) and SleuthKit open source forensics frameworks, respectively, and, accordingly, the proposed framework is evaluated in a forensically sound manner. The evaluation result is compared with other relevant works and, as a result, the proposed solution provides a privacy-preserving, efficient forensic imaging and analysis process while having also sufficient methods. Moreover, the AFF4 forensic image, produced by the SIM module, can be analyzed not only by SAM, but also by other well-known analysis tools available on the market.

Deep Feature Enhancement Method for Land Cover With Irregular and Sparse Spatial Distribution Features: A Case Study on Open-Pit Mining

Land cover classification in mining areas (LCMA) is essential for the environmental assessment of mines and plays a crucial role in their sustainable development. The shapes of mine land occupation elements are irregular, and the overall proportion of their area is relatively small. Therefore, their features may be easily lost during feature extraction, which limits the interpretation accuracy in mining areas. This study attempts to address these issues. We propose a model named EG-UNet to enhance the features of elements with few samples and to capture long-range information. The proposed EG-UNet includes two main modules: 1) The edge feature enhancement module: the edges of elements of mine land occupation contain more information than other spatial locations. Hence, during the feature extraction of elements, a Sobel operator is used to extract the object boundary, which increases the weight of these features before the pooling operation for their preservation. 2) The long-range information extraction module: long-range information helps extract tiny objects, such as dumping grounds in the mining area. We present a graph convolution network (GCN) to capture the long-range features and apply convolutional neural networks to learn the graph construction. A total of 10 deep-learning networks were compared using the LCMA semantic segmentation dataset. Our model exhibited the best performance, especially in classifying classes with few samples. Further, to evaluate the general ability of EG-UNet, a benchmark-Gaofen Image Dataset (GID) was used, and the result still reflected the superiority of our method.

COVID-19 disease identification network based on weakly supervised feature selection.

The coronavirus disease 2019 (COVID-19) outbreak has resulted in countless infections and deaths worldwide, posing increasing challenges for the health care system. The use of artificial intelligence to assist in diagnosis not only had a high accuracy rate but also saved time and effort in the sudden outbreak phase with the lack of doctors and medical equipment. This study aimed to propose a weakly supervised COVID-19 classification network (W-COVNet). This network was divided into three main modules: weakly supervised feature selection module (W-FS), deep learning bilinear feature fusion module (DBFF) and Grad-CAM++ based network visualization module (Grad-Ⅴ). The first module, W-FS, mainly removed redundant background features from computed tomography (CT) images, performed feature selection and retained core feature regions. The second module, DBFF, mainly used two symmetric networks to extract different features and thus obtain rich complementary features. The third module, Grad-Ⅴ, allowed the visualization of lesions in unlabeled images. A fivefold cross-validation experiment showed an average classification accuracy of 85.3%, and a comparison with seven advanced classification models showed that our proposed network had a better performance.

Electrical Impedance Tomography Hardware with Demodulation

Electrical impedance tomography (EIT) is a promising imaging technology. It is portable, user-independent, and low cost (compared to other imaging technologies). There are two main approaches for EIT equipments: dynamic and absolute images. Absolute images are technologically more difficult and are limited by hardware. Phase information is required to create permittivity images. This is an ongoing research with the objective of evaluating the phase information. A new architecture with up to 64 electrodes is proposed. Each electrode has its own Howland current source and microcontroller. The proposed hardware consists of three main modules: the electrode hardware (with Howland current source and microcontroller); the demodulator algorithm; and, the electrode synchronizer. Some results are presented with the electrode hardware that demonstrate data acquisition with two electrodes. Tests with the demodulator algorithm were also presented. Simulated tests with the designed electrode synchronizer were performed, and the implementation remains, and it is considered a future work.

CVML-Pose: Convolutional VAE Based Multi-Level Network for Object 3D Pose Estimation

Most vision-based 3D pose estimation approaches typically rely on knowledge of object’s 3D model, depth measurements, and often require time-consuming iterative refinement to improve accuracy. However, these can be seen as limiting factors for broader real-life applications. The main motivation for this paper is to address these limitations. To solve this, a novel Convolutional Variational Auto-Encoder based Multi-Level Network for object 3D pose estimation (CVML-Pose) method is proposed. Unlike most other methods, the proposed CVML-Pose implicitly learns an object’s 3D pose from only RGB images encoded in its latent space without knowing the object’s 3D model, depth information, or performing a post-refinement. CVML-Pose consists of two main modules: (i) CVML-AE representing convolutional variational autoencoder, whose role is to extract features from RGB images, (ii) Multi-Layer Perceptron and K-Nearest Neighbor regressors mapping the latent variables to object 3D pose including, respectively, rotation and translation. The proposed CVML-Pose has been evaluated on the LineMod and LineMod-Occlusion benchmark datasets. It has been shown to outperform other methods based on latent representations and achieves comparable results to the state-of-the-art, but without use of a 3D model or depth measurements. Utilizing the t-Distributed Stochastic Neighbor Embedding algorithm, the CVML-Pose latent space is shown to successfully represent objects’ category and topology. This opens up a prospect of integrated estimation of pose and other attributes (possibly also including surface finish or shape variations), which, with real-time processing due to the absence of iterative refinement, can facilitate various robotic applications. Code available: https://github.com/JZhao12/CVML-Pose.

Consensus graph learning-based self-supervised ensemble clustering

目的 随着实际应用场景中海量数据采集技术的发展和数据标注成本的不断增加，自监督学习成为海量数据分析的一个重要策略。然而，如何从海量数据中抽取有用的监督信息，并该监督信息下开展有效的学习仍然是制约该方向发展的研究难点。为此，提出了一个基于共识图学习的自监督集成聚类框架。方法 框架主要包括3个功能模块。首先，利用集成学习中多个基学习器构建共识图；其次，利用图神经网络分析共识图，捕获节点优化表示和节点的聚类结构，并从聚类中挑选高置信度的节点子集及对应的类标签生成监督信息；再次，在此标签监督下，联合其他无标注样本更新集成成员基学习器。交替迭代上述功能块，最终提高无监督聚类的性能。结果 为验证该框架的有效性，在标准数据集（包括图像和文本数据）上设计了一系列实验。实验结果表明，所提方法在性能上一致优于现有聚类方法。尤其是在MNIST-Test （modified national institute of standards and technology database）上，本文方法实现了97.78%的准确率，比已有最佳方法高出3.85%。结论 该方法旨在利用图表示学习提升自监督学习中监督信息捕获的能力，监督信息的有效获取进一步强化了集成学习中成员构建的能力，最终提升了无监督海量数据本质结构的挖掘性能。;Objective Clustering is focused on machine learning-related data segmentation for multiple datasets. Its applications are in relevant to such domains like image segmentation and anomaly detection. In addition，to simplify complex tasks optimize its performance，clustering is used in data preprocessing tasks of those are data sub-blocks segmentation， pseudo-labels generation，and abnormal points-removal. Self-supervised learning has become an essential technique for massive data analysis. However，it is challenged to extract effective supervision information and analyze the input data. Method A consensus graph learning based self-supervised ensemble clustering（CGL-SEC）framework is developed. It consists of three main modules：1）to construct the consensus graph based on several ensemble components（i. e. ，the basic clustering methods）. 2）to extract the supervision information by learning the consensus graph representation，and 3）its node clustering results，where the subset of nodes with the high-confidence are selected as labeled samples. To optimize the ensemble components and the corresponding consensus graph，t basic clustering methods are re-trained in related the option of samples-labeled and other samples-unlabeled. The final clustering results can be optimized iteratively until the learning process converges. Result A series of experiments are carried out on benchmarks，including both image and textual datasets. Especially，CGL-SEC is 3. 85% over baseline in terms of clustering evaluation metric on themodified national institute of standards and technology database（MNIST-Test）. First，to optimize data representation and cluster assignment at the same time，deep embedding clustering can be focused on data itself as the supervision information and auto-encoder with the reconstruction loss is pre-trained. The soft cluster assignment of features-embedded is then calculated，and the KL（Kullback-Leibler）divergence is minimized between the soft cluster assignment and the auxiliary target distribution. To improve the performance of the model further，following deep clustering network（DCN）can use hard clustering instead of soft allocation，and local constraints are applied by improved deep embedding clustering（IDEC）. The pseudo-label strategy is implemented as a self-supervised learning method that uses the prediction results of the neural network as the label to simulate the supervision information compared to using data itself as the supervision information. Deepcluster-based K-means clustering is used to generate pseudo-labels to guide the training of convolutional networks. However，the generated pseudo-labels have lower confidence and are prone to trivial solutions in the initial stage of network training. Deep embedding clustering with data augmentation（DEC-DA）and MixMatch-based prediction of data-enhanced samples are used as the supervision information of the original data，which improves the accuracy of the supervision information to a certain extent，but this method is difficult to extend to text and other fields. Deep adaptive clustering-based high-confidence pseudo-label subsets-selected are iteratively trained the network in the prediction results，but lowconfidence samples-involved data distribution information is ignored. Pseudo-semi-supervised clustering votes are used to select a subset of high-confidence pseudo-labels，and all samples are used to train semi-supervised neural network. Although the ensemble strategy can improve the confidence of the pseudo-label，the voting strategy is concerned of category representation only without the feature representation of the sample itself，which can reduce the clustering performance in some cases. The ensemble learning is regarded as a representative machine learning method that reflects the ability of group intelligence，whereas a learning method can improve the overall prediction performance via multiple base learners training and their coordinated prediction results. In pseudo-label-based clustering tasks，it can coordinate multiple base learners to obtain high-confidence pseudo-labels. However，the effectiveness of the supervision information acquisition is still to be resolved. The category information of the sample is considered for current pseudo-label-based ensemble clustering method only when the label is captured and some effective information are ignored like the feature representation of the sample itself and the clustering structure between samples. Conclusion Graph neural network is composed of content information of nodes and the structural information between nodes at the same time. To design a self-supervised ensemble clustering method based on consensus graph representation learning，it is required to make full use of sample features and relationships between samples in ensemble learning. To obtain higher confidence pseudo-labels as supervised information and improve the performance of self-supervised clustering，it is necessary to mine global and local information at the same time. We illustrate a learnable data ensemble representation through graph neural network. The confidence of pseudo-labels is improved，and the entire model is trained in self-supervision iteratively. To be summarized：1）Commonly-used consensus graph learning-integrated clustering framework is developed，which can use multi-level information like clusteringintegrated sample characteristics and category structure. 2）Self-supervision method is proposed，which uses graph neural network to mine the global and local information of the consensus graph，and high-confidence pseudo-labels are obtained as supervised information. 3）Experiments are demonstrated that the consensus graph learning ensemble clustering method has its potentials on image and text datasets.

DEVELOPMENT OF METHODOLOGICAL SUPPORT ON MOLECULAR BIOLOGY FOR MULTILINGUAL SECONDARY SCHOOL CLASSES

The aim of the work is to develop a multilingual teaching and learning aid on molecular biology for schools and universities. The algorithm of creating tasks on molecular biology was used, including determining the type or structure of the problem, corresponding to the life situation, formulation of the circumstance and problem question. Approbation work was carried out in school-gymnasium No. 2 named after Gafu Kayyrbekov. in Astana. On the basis of the ascertaining experiment, 60% of children or 40 students are defined as having a low level of performance in the fulfillment of tasks in molecular biology. About 25% of all respondents were able to solve tasks correctly. Only 10 people or 15% of respondents coped with the tasks at a high level. In the course of the experiment we used the tasks on molecular biology developed by the authors, divided into four main modules, each subtopic of the module contains learning objectives, content, animation and tasks. After the work done, the share of students who showed a low level of fulfillment up to 20% . The average level of fulfillment of tasks in molecular biology during the observation period was recorded in 32 students (50%). And the indicators at the high level increased 2 times and amounted to 30%.

Structure-Informed Shadow Removal Networks.

Existing deep learning-based shadow removal methods still produce images with shadow remnants. These shadow remnants typically exist in homogeneous regions with low-intensity values, making them untraceable in the existing image-to-image mapping paradigm. We observe that shadows mainly degrade images at the image-structure level (in which humans perceive object shapes and continuous colors). Hence, in this paper, we propose to remove shadows at the image structure level. Based on this idea, we propose a novel structure-informed shadow removal network (StructNet) to leverage the image-structure information to address the shadow remnant problem. Specifically, StructNet first reconstructs the structure information of the input image without shadows and then uses the restored shadow-free structure prior to guiding the image-level shadow removal. StructNet contains two main novel modules: (1) a mask-guided shadow-free extraction (MSFE) module to extract image structural features in a non-shadow-to-shadow directional manner, and (2) a multi-scale feature & residual aggregation (MFRA) module to leverage the shadow-free structure information to regularize feature consistency. In addition, we also propose to extend StructNet to exploit multi-level structure information (MStructNet), to further boost the shadow removal performance with minimum computational overheads. Extensive experiments on three shadow removal benchmarks demonstrate that our method outperforms existing shadow removal methods, and our StructNet can be integrated with existing methods to improve them further.

ЧИСЛЕННОЕ ИССЛЕДОВАНИЕ ДИНАМИКИ ВНУТРИКАМЕРНЫХ ПРОЦЕССОВ В МНОГОСОПЛОВОМ РАКЕТНОМ ДВИГАТЕЛЕ НА ТВЕРДОМ ТОПЛИВЕ ЧАСТЬ 1. МЕТОДИКА РАСЧЁТА

The dynamics of transient in-chamber processes of booster (overclocking) multi-hop of solid rocket motor is investigated. The research method is the formulation of a computational experiment. The conjugate formulation of the problem is considered, which includes: – the ignition device is triggered (the combustion rate of the igniting composition is described on the basis of an experimental and theoretical approach that takes into account the burning of combustion products behind the body of the igniting device); – heating, ignition and subsequent unsteady and turbulent combustion of a charge of a mixed solid fuel (a quasihomogeneous combustion model is used, formulated on the basis of the equations of thermal conductivity and chemical kinetics recorded for the condensed phase (solid fuel), taking into account the influence of the gas phase (flare) on the combustion process in the condensed phase; the method of solving the problem is the finite difference method); – unsteady three-dimensional homogeneous-heterogeneous three-phase flow of nitrogen (pre-pressurization gas of the combustion chamber) and combustion products of the igniting composition and mixed solid fuel in the combustion chamber, multi-core block and behind the multi-core block of the rocket engine (approaches of mechanics of continuous multiphase media are used; the basic system of equations is a system of vortex differential equations of gas dynamics; the solution method is multiparametric a class of difference splitting schemes by physical processes of the Davydov method); – depressurization of the combustion chamber and the departure of the plugs of the multi-nozzle block of solid rocket motor (the equation of motion of the nozzle block plug is Newton's second law; the method of solving the problem is the Euler method). Each of the subtasks is considered in relation and resolved simultaneously – at one-time step. To solve the formulated problem, a set of application programs has been developed using (for the main calculation module) the OpenCL multithreading information processing standard. The operability of the software product was checked.

The Blue-RoSES Project: a Gate for Remote Exploration of the Seas

This paper focuses on the description of the Blue-RoSES project, an European funded activity devoted to the development and exploitation of a reliable infrastructure for the remote employment of Remotely Operated Vehicles (ROVs) for both professional and recreational applications in coastal environments, marinas and yachting. The paper firstly provides an overall view of the concept and then summarizes the main modules composing the architecture that is responsible for reliable and stable remote piloting of ROVs. Analysis of the critical issues related to remote piloting and control is carried out and measurements of the communication lags and delays are performed to obtain a statistical characterization, also providing a quantitative measure of the communication service. The reliability and performance of the complete system are demonstrated through experimental campaigns in real case scenarios.

Fenotipo de la enfermedad de células falciforme. Relación de los haplotipos y los polimorfismos en el clúster β, BCL11A y HBS1L-MYB. Estudio piloto

Sickle cell disease (SCD), despite being a monogenic disease, presents a highly variable phenotype that essentially depends on the amount of fetal hemoglobin (HbF), constituting the main modulator of the disease. The variation of HbF levels between patients is genetically regulated. HbF determines both the phenotype of the disease and the response to treatment with the main drug used, hydroxyurea. Researchers’ efforts have focused on discovering the genetic factors responsible for its variation; mainly describing the haplotypes of the β cluster and SNPs in three different loci: BCL11A, HBS1L-MYB and the β-globin cluster. Objective: To determine, in a cohort of patients with SCD, the possible relationship between the number of SNPs and haplotypes with higher levels of HbF. A positive association could explain why certain haplotypes, such as Senegal or Arab-Indian, have higher levels. of HbF and less severe disease. Methods: To verify this hypothesis, the characterization of the haplotypes was carried out using the PCR-RFLP technique and the genotyping of three SNPs representative of the three loci with the greatest association with the variation of HbF: XmnI (rs7482144), BCL11A (rs4671393) and HBS1L-MYB (rs9376092). Results: A greater number of SNPs has been reported in the haplotypes related to higher HbF and a lower number in those with less HbF, although only the SNP XmnI (rs7482144) has shown a statistically significant association. Conclusions: A direct relationship between haplotypes and the number of SNPs has been found, reporting that haplotypes with higher levels of HbF and thus a less severe phenotype, had a greater number of SNPs. Thus, the Benín and Bantu haplotypes traditionally associated with poor prognosis are the ones that have shown a lower number of mutated SNPs.

Landmark Recognition Model for Smart Tourism using Lightweight Deep Learning and Linear Discriminant Analysis

Scene recognition algorithm is crucial for landmark recognition model development. Landmark recognition model is one of the main modules in the intelligent tour guide system architecture for the use of smart tourism industry. However, recognizing the tourist landmarks in the public places are challenging due to the common structure and the complexity of scene objects such as building, monuments and parks. Hence, this study proposes a super lightweight and robust landmark recognition model by using the combination of Convolutional Neural Network (CNN) and Linear Discriminant Analysis (LDA) approaches. The landmark recognition model was evaluated by using several pretrained CNN architectures for feature extraction. Then, several feature selections and machine learning algorithms were also evaluated to produce a super lightweight and robust landmark recognition model. The evaluations were performed on UMS landmark dataset and Scene-15 dataset. The results from the experiments have found that the Efficient Net (EFFNET) with CNN classifier are the best feature extraction and classifier. EFFNET-CNN achieved 100% and 94.26% classification accuracy on UMS-Scene and Scene-15 dataset respectively. Moreover, the feature dimensions created by EFFNet are more compact compared to the other features and even have significantly reduced for more than 90% by using Linear Discriminant Analysis (LDA) without jeopardizing classification performance but yet improved its performance.

A Novel Framework for Semi-supervised Multiple-label Image Classification using Multi-stage CNN and Visual Attention Mechanism

To train deep neural networks effectively, a lot of labeled data is typically needed. However, real-time applications make it difficult and expensive to acquire high-quality labels for the data because it takes skill and knowledge to accurately annotate multiple label images. In order to enhance classification performance, it is also crucial to extract image features from all potential objects of various sizes as well as the relationships between labels of numerous label images. The current approaches fall short in their ability to map the label dependencies and effectively classify the labels. They also perform poor to label the unlabeled images when small amount of labeled images available for classification. In order to solve these issues, we suggest a new framework for semi-supervised multiple object label classification using multi-stage Convolutional neural networks with visual attention (MSCNN)and GCN for label co-occurrence embedding(LCE) (MSCNN-LCE-MIC), which combines GCN and attention mechanism to concurrently capture local and global label dependencies throughout the entire image classification process. Four main modules make up MSCNN-LCE-MIC: (1) improved multi-label propagation method for labeling largely available unlabeled image; (2) a feature extraction module using multi-stage CNN with visual attention mechanism that focuses on the connections between labels and target regions to extract accurate features from each input image; (3) a label co-existence learning that applies GCN to discover the associations between different items to create embeddings of label co-occurrence; and (4) an integrated multi-modal fusion module. Numerous tests on MS-COCO and PASCAL VOC2007 show that MSCNN-LCE-MIC significantly improves classification efficiency on mAP 84.3% and 95.8% respectively when compared to the most recent existing methods.

Main modulators of COVID-19 epidemic in sub-Saharan Africa

BackgroundThe coronavirus disease 2019 (COVID-19) pandemic is responsible for an important global death toll from which sub-Saharan Africa (SSA) seems mostly protected. The reasons explaining this situation are still poorly understood. MethodsWe analyzed the correlation between reported COVID-19 data between February 14, 2020 and May 18, 2021, and demographic, socioeconomic, climatic, diagnostic data, and comorbidities in 47 SSA countries. Different databases including the WHO data center, Our World in Data, and the World Bank were used. FindingsAs of May 17, 2021, SSA reported 2% of COVID-19 cases and 2.9% of deaths, with the southern region being the most affected with 56.4% of cases and 75.0% of deaths. COVID-19 mortality was positively correlated with medical variables (national obesity rate, diabetes prevalence, cancer incidence, and cardiovascular disease mortality rate), socioeconomic characteristics (international tourism, per capita health expenditure, human development index, HDI, and years of schooling), and health system variables (nurse density, number of COVID-19 tests per capita), but negatively correlated with the population under 15 years of age and the malaria index. InterpretationOur study suggests that higher economic status fits with high COVID-19 mortality in SSA. In this regard, it represents primarily a disease of modern and wealthy societies, and can therefore be considered as an exception among infectious diseases that historically affected more severely underserved populations living in low- and middle-income countries. However, it should be made clear that observed correlations do not imply inevitably causation and that additional studies are necessary to confirm our observations.

The Historical Statistics of the Altai Information System

In the era of the digital turn, it is becoming relevant to create information resources that provide researchers with not only background information about historical sources but also reliable historical data. At the all-Russian level, such resources come from library centres (National Electronic Library, Presidential Library), archival authorities (Central Fund Catalogue), and public organisations (Russian Historical Society, Russian Geographical Society). For some regions, this task remains new and requires modern approaches and solutions. The article provides a description of the information system for a set of statistical materials on the history of Altai in the nineteenth – the first half of the twentieth centuries kept in central and regional libraries, archives, digitised publications, and collections with the possibility of versatile search and analysis of historical data. On the one hand, the chronological boundaries of the materials included in the database depend on the formation of the foundations of regular statistics in the mid-nineteenth century, and, on the other hand, on the establishment of Altai Krai within its present-day boundaries in 1937. The territorial framework of the study covers the borders of Altai as a historical region, which was included in the period under review in such administrative-territorial formations as Tomsk and Altai provinces and Siberian and West Siberian regions. The study employs a systematic approach, examining various types of statistical data as elements of a single set of sources on the history of Altai in the nineteenth – first half of the twentieth centuries. In its IT aspect, the study relies on database technology, both a tool and a result of the study. The Historical Statistics of Altai information system was created under the control of MySQL and consists of directories and four main modules: storage, sources, tags, and documents. The authors examine database architecture, logical structure, and data characteristics of the modules. The information system on the historical statistics of Altai will make it possible to solve some of the heuristic tasks related to the search and use of archival, museum, and library funds, including familiarisation with historical sources and their preliminary analysis in remote access mode. As a result, it will significantly expand research opportunities.