The study compares three machine learning and deep learning models—Support Vector Regression (SVR), Recurrent Neural Networks (RNN), and Long Short-Term Memory (LSTM)—for predicting market prices using the NGX All-Share Index dataset. The models were evaluated using multiple error metrics, including Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Square Error (RMSE), Mean Percentage Error (MPE), and R-squared. RNN and LSTM were tested with both 30 and 60-day windows, with performance compared to SVR. LSTM delivered better R-squared values, with a 60-day LSTM achieving the best accuracy (R-squared = 0.993) when using a combination of endogenous market data and technical indicators. SVR showed reliable results in certain scenarios but struggled in fold 2 with sudden spike that shows high probability of not capturing the entire underlying NGX pattern in the dataset correctly as witnessed with the high validation loss during the period. Additionally, RNN faced the vanishing gradient problem that limits its long-term performance. Despite challenges, LSTM's ability to handle temporal dependencies, especially with the inclusion of On-Balance Volume, led to significant improvements in prediction accuracy. The use of the Optuna optimization framework further enhanced model training and hyperparameter tuning, contributing to the performance of the LSTM model.

Network Function Virtualization (NFV) allows for the dynamic provisioning of Virtualized Network Functions, adapting services to the complex and real-time network environment for enhanced network performance. Despite the benefits of NFV, virtual network functions (VNFs) migration and the energy consumption pose significant challenges due to the dynamic nature of the physical network, especially in the edge-cloud continuum network. Currently, people tackle this challenge by predicting the network traffic and then proactively migrating the VNFs using the predicted values. There are two challenges here. First, when SFCs belong to different network service providers, the historical network traffic data is held by the service providers, and they are reluctant to share the historical data due to privacy concerns. Without the historical network traffic data of each SFC, the resource provider that owns the underlying network cannot effectively predict the network traffic; in addition, without the feedback of the migration results of the underlying physical network, the network traffic prediction will be difficult to match the needs of the physical network. In this manuscript, we solve the problem of network traffic prediction in the NFV-based edge-cloud continuum network for multiple network service providers. We aim to improve migration performance and reduce energy consumption while ensuring resource and latency requirements. To protect the privacy of each network service provider, we apply an FL algorithm. After that, to improve the migration performance, we tailor the loss function. In the loss function, factors such as the number of migrations, the number of migration failures, and the energy consumption of the NFV-based edge-cloud continuum network are added. Additionally, to obtain the results of the SFC migration, we develop a mathematical model for the problem of multiple network service providers in the NFV-based network. The effectiveness of our algorithm is evaluated through extensive simulations, and the results show a significant reduction in the number of migrated nodes and energy consumption while improving the SFC acceptance ratio compared to state-of-the-art schemes which only use the difference between predicted and actual flows to define the loss function.

We study latching dynamics in the adaptive Potts model network, through numerical simulations with randomly and also weakly correlated patterns, and we focus on comparing its slowly and fast adapting regimes. A measure, Q, is used to quantify the quality of latching in the phase space spanned by the number of Potts states S, the number of connections per Potts unit C and the number of stored memory patterns p. We find narrow regions, or bands in phase space, where distinct pattern retrieval and duration of latching combine to yield the highest values of Q. The bands are confined by the storage capacity curve, for large p, and by the onset of finite latching, for low p. Inside the band, in the slowly adapting regime, we observe complex structured dynamics, with transitions at high crossover between correlated memory patterns; while away from the band latching transitions lose complexity in different ways: below, they are clear-cut but last so few steps as to span a transition matrix between states with few asymmetrical entries and limited entropy; while above, they tend to become random, with large entropy and bi-directional transition frequencies, but indistinguishable from noise. Extrapolating from the simulations, the band appears to scale almost quadratically in the p - S plane, and sublinearly in p - C. In the fast adapting regime the band scales similarly, and it can be made even wider and more robust, but transitions between anti-correlated patterns dominate latching dynamics. This suggest that slow and fast adaptation have to be integrated in a scenario for viable latching in a cortical system. The results for the slowly adapting regime, obtained with randomly correlated patterns, remain valid also for the case with correlated patterns, with just a simple shift in phase space.

Deep learning methods have been widely used in PolSAR image classification. To learn the polarimetric information, many deep learning methods expect to learn high-level semantic features from original PolSAR data. However, only original data cannot learn multiple scattering features and complex structures for extremely heterogeneous terrain objects. In addition, deep learning methods always cause edge confusion due to the high-level features. To overcome these shortages, we propose a double-channel CNN network combined with an edge-preserving MRF model(DCCNN-MRF) for PolSAR image classification. Firstly, to combine complex matrix data and multiple scattering features together, a double-channel convolution network(DCCNN) is developed, which consists of a Wishart-based complex matrix and multi-feature subnetworks. The Wishart-based complex matrix network can learn the statistical characteristics and channel correlation well, and the multi-feature network can learn high-level semantic features well. Then, a unified network framework is designed to fuse two kinds of features to enhance advantageous features and reduce redundant ones. Finally, an edge preserving MRF model is designed to combine with the DCCNN network. In the MRF model, a sketch map-based edge energy function is designed by defining adaptive weighted neighborhood for edge pixels. Experiments are conducted on four real PolSAR data sets with different sensors and bands. Experimental results demonstrate the effectiveness of the proposed DCCNN-MRF method.

Link prediction is to complete the missing links in the network or to predict the generation of new links according to the current network structure information, which is very important for mining and analyzing the evolution of the network such for construction and analysis of logical architecture in 6G network. Link prediction algorithms based on node similarity need predefined similarity functions, which is highly hypothetical and only applies to specific network structures without generality. To solve this problem, this paper proposes a link prediction algorithm based on the subgraph of the target node pair. In order to automatically learn the graph structure characteristics, the algorithm firstly extracts the h-hop subgraph of the target node pair, and then predicts whether the target node pair will be linked according to the subgraph. Experiments on seven real data sets show that the link prediction algorithm based on target node pair subgraph is suitable for various network structures and superior to other link prediction algorithms.

As V2X technology develops, acute problems related to reliable and secure information exchange between network objects in real time appear. The article aims to solve the scientific problem of building a network architecture for reliable delivery of correct and uncompromised data within the V2X concept to improve the safety of road users, using blockchain technology and mobile edge computing (MEC). The authors present a formalized mathematical model of the system, taking into account the interconnection of objects and V2X information channels, and an energy-efficient algorithm of traffic offloading to the MEC server. The paper presents the results of application of blockchain technologies and mobile edge computing in the developed system, their description, evaluation of advantages and disadvantages of the implementation.

The implementation of neural networks (NN) on edge devices enables local processing of wireless data but faces challenges such as high computational complexity and memory requirements when deep neural networks (DNN) are used. Shallow neural networks customized for specific problems are more efficient, requiring fewer resources, and resulting in a lower latency solution. An additional benefit of the smaller network size is that it is suitable for real-time processing on edge devices. The main concern with shallow neural networks is their accuracy performance compared to DNNs. In this paper, we demonstrate that a customized adaptive activation function (AAF) can meet the accuracy of a DNN. We designed an efficient FPGA implementation for a customized segmented spline curve neural network (SSCNN) structure to replace the traditional fixed activation function with an AAF. We compared our SSCNN with different neural network structures such as real-valued time delay neural network (RVTDNN), augmented real-valued time delay neural network (ARVTDNN), and deep neural networks with different parameters. Our proposed SSCNN implementation uses 40% fewer hardware resources and no Block RAMS compared to the DNN with similar accuracy. We experimentally validate this computationally efficient and memory-saving FPGA implementation of SSCNN for digital predistortion of RF power amplifiers using the AMD/Xilinx RFSoC ZCU111. The implemented solution uses less than 3% of the available resources. The solution also enables an increase of the clock frequency to 221.12 MHz, allowing the transmission of wide bandwidth signals.

Chickpea is one of the most widely consumed pulses globally because of its high protein content. The morphological features of chickpea seed, such as colour, texture, are observable and play a major role in classifying different chickpea varieties. This process is often carried out by human experts, and is time-consuming, inaccurate, and expensive. The objective of the study was to design an automated chickpea classifier using an RGB colour image-based model by considering the morphological features of chickpea seed. As part of the data acquisition process, five hundred and fifty images were collected per variety for four varieties of chickpea (CDC-Alma, CDC-Consul, CDC-Cory, and CDC-Orion) using an industrial RGB camera and a mobile phone camera. Three CNN-based models such as NasNet-A (mobile), MobileNetV3 (small), and EfficientNetB0 were evaluated using a transfer learning-based approach. The classification accuracy was 97%, 99%, and 98% for NasNet-A (mobile), MobileNetV3 (small), and EfficientNetB0 models, respectively. The MobileNetV3 model was used for further deployment on an Android mobile and Raspberry Pi 4 devices based on its higher accuracy and light-weight architecture. The classification accuracy for the four chickpea varieties was 100% while the MobileNetV3 model was deployed on both Android mobile and Raspberry Pi 4 platforms.

Over the past few years, the importance of video surveillance in securing the national critical infrastructure has significantly increased, whose applications include detecting failures and anomalies. Accompanied by video proliferation is the increasing number of attacks against surveillance systems. Among the attacks, false frame injection (FFI) attacks that replay video frames from a previous recording to mask the live feed has the highest impact. While many attempts have been made to detect FFI frames using features from the video feeds, video analysis is computationally too intensive to be deployed on-site for real-time false frame detection. In this paper, we investigate the feasibility of FFI attacks on compromised surveillance systems at the edge and propose an effective technique to detect the injected false video and audio frames by monitoring the surveillance feed using the embedded Electrical Network Frequency (ENF) signals. An ENF operates at a nominal frequency of 60 Hz/50 Hz based on its geographical location and maintains a stable value across the entire power grid interconnection with minor fluctuations. For surveillance system video/audio recordings connected to the power grid, the ENF signals are embedded. The time-varying nature of the ENF component is used as a forensic application for authenticating the surveillance feed. The paper highlights the ENF signal collection from a power grid creating a reference database and ENF extraction from the recordings using conventional short-time Fourier Transform and spectrum detection for robust ENF signal analysis in the presence of noise and interference caused in different harmonics. The experimental results demonstrate the effectiveness of ENF signal detection and/or abnormalities for FFI attacks.

As an emerging network paradigm, Space-Air-Ground integrated networks (SAGIN) has attracted the attentions from academia and industry. That is because SAGIN can implement seamless global coverage and connections among electronic devices in space, air, and ground spaces. Additionally, the shortages of computing and storage resources in mobile devices greatly affect the quality of experiences for intelligent applications. Hence, we devise to integrate SAGIN as an abundant resource pool into mobile edge computing environments (MEC). To facilitate efficient processing, we need to solve the optimal task offloading decisions. Different from the existing MEC task offloading solutions, we have to face some new challenges, such as the fluctuation of processing capability for an edge computing node, the uncertainty of the transmission latency caused by the heterogeneous network protocols, the uncertain amount of uploaded tasks during a period, and so on. In this paper, we firstly describe a task offloading decision problem in new challenge environments. But, we cannot use the standard robust optimization and stochastic optimization methods to obtain the optimal result under the uncertain network environments. In this paper, we propose the condition value at risk-aware distributionally robust optimization algorithm, named as CVAR-DRO, to solve the task offloading decision problem. The proposed CVAR-DRO method combines the distributionally robust optimization and the condition value at risk model for solving the optimal result. And then, We have evaluated our approach in simulation SAGIN environments with the confidence interval, the number of mobile task-offloading and the various parameters. We compare our proposed CVAR-DRO algorithm with the state-of-the-art algorithms, such as the standard robust optimization algorithm, the stochastic optimization algorithm, the DRO algorithm, and the brute algorithm. The experimental results show that CVAR-DRO can get a sub-optimal mobile task-offloading decision. Overall, CVAR-DRO is more robust than others to the new challenges mentioned above in SAGIN.

The detection performance of small objects in remote sensing images has not been satisfactory compared to large objects, especially in low-resolution and noisy images. A generative adversarial network (GAN)-based model called enhanced super-resolution GAN (ESRGAN) showed remarkable image enhancement performance, but reconstructed images usually miss high-frequency edge information. Therefore, object detection performance showed degradation for small objects on recovered noisy and low-resolution remote sensing images. Inspired by the success of edge enhanced GAN (EEGAN) and ESRGAN, we applied a new edge-enhanced super-resolution GAN (EESRGAN) to improve the quality of remote sensing images and used different detector networks in an end-to-end manner where detector loss was backpropagated into the EESRGAN to improve the detection performance. We proposed an architecture with three components: ESRGAN, EEN, and Detection network. We used residual-in-residual dense blocks (RRDB) for both the ESRGAN and EEN, and for the detector network, we used a faster region-based convolutional network (FRCNN) (two-stage detector) and a single-shot multibox detector (SSD) (one stage detector). Extensive experiments on a public (car overhead with context) dataset and another self-assembled (oil and gas storage tank) satellite dataset showed superior performance of our method compared to the standalone state-of-the-art object detectors.

The Internet-of-Things (IoT) triggers new types of cyber risks. Therefore, the integration of new IoT devices and services requires a self-assessment of IoT cyber security posture. By security posture this article refers to the cybersecurity strength of an organisation to predict, prevent and respond to cyberthreats. At present, there is a gap in the state-of-the-art, because there are no self-assessment methods for quantifying IoT cyber risk posture. To address this gap, an empirical analysis is performed of 12 cyber risk assessment approaches. The results and the main findings from the analysis is presented as the current and a target risk state for IoT systems, followed by conclusions and recommendations on a transformation roadmap, describing how IoT systems can achieve the target state with a new goal-oriented dependency model. By target state, we refer to the cyber security target that matches the generic security requirements of an organisation. The research paper studies and adapts four alternatives for IoT risk assessment and identifies the goal-oriented dependency modelling as a dominant approach among the risk assessment models studied. The new goal-oriented dependency model in this article enables the assessment of uncontrollable risk states in complex IoT systems and can be used for a quantitative self-assessment of IoT cyber risk posture.

The forthcoming paradigms of 6G, Future Networks (FN) and Super-Internet of Things (IoT), will bring disruption at various levels of the physical infrastructure. This work focuses on the network edge, stressing how the forecasted proliferation and technology innovation of the physical systems at the boundary of the network (edge), will pose crucial issues to be addressed in the next years. In particular, the continuity from the edge to the core of the network will be addressed, suggesting unprecedented design and development approaches. Along such a direction, the impact of Micro and Nanotechnologies as a Key Enabling Technology (KET) for the network edge of the future is sketched, including numerous examples of already existing micro/nano devices, components and systems. Afterwards, a practical study of how improving hardware technologies at the edge can be beneficial in terms of more efficient operation is reported. To this end, scattered intelligence shall be a key enabler with federated learning at the base station providing decisive assistance to adaptive downlink beamforming codebook design for mmWave massive MIMO set-ups as essential ingredients for enhanced link reliability in edge networks.

This paper outlines a tool developed with the purpose of creating a simple configurable emulated network environment that can be used in cyber defence exercises. Research has been conducted into the various related subject areas: cyber defence exercises, network threats, network emulation, network traffic replay, network topologies, and common network services. From this research a requirements specification was produced to encapsulate the features required to create this tool. A network, containing many of the aspects researched, was designed and implemented using Netkit-NG to act as a blueprint for the tool and to further knowledge in the construction of an emulated network. Following this the tool was developed and tested to ensure requirements were met.

: Biological networks such as protein interaction networks, gene regulation networks and metabolic pathways are examples of complex networks which are large graphs with small-world and scale-free properties. Analysis of these networks has a profound effect on our understanding the origins of life, health and disease states of organisms, and diagnose diseases to aid the search for remedial processes. In this review, we describe main analysis methods of biological networks using graph theory by first defining main parameters such as clustering coefficient, modularity and centrality. We then survey fundamental graph clustering methods and algorithms followed by the network motif search algorithms with the aim of finding repeating subgraphs in a biological network graph. A frequently appearing subgraph usually conveys a basic function carried out by that small network and discovering such a function provides an insight to the overall function of the organism. Lastly, we review network alignment algorithms that achieve to find similarities between two or more graphs representing biological networks. A conserved subgraph between the biological networks of organisms may mean a common ancestor and finding such relationship may help researchers derive ancestral relationships and predict the future evolution of organisms to enable designing new drugs. We conclude by the current challenging areas of biological network analysis using graph theory and parallel processing for high performance analysis

The article presents the Base Point Split (BPSplit) algorithm for generating a polygon skeleton based on sets of vector data describing lakes and rivers. A key feature of the BPSplit algorithm is that it is dependent on base points representing the source or mouth of a river or a stream. The input values of base points determine the shape of the resulting skeleton. Various skeletons can be generated with the use of different base points. Base points are applied to divide polygon boundaries into segments. Segmentation supports the selection of TIN edges inside polygons. The midpoints of selected TIN edges constitute a basis for generating a skeleton. The algorithm handles polygons with numerous holes, and it accounts for all holes. This article proposes a method for modifying a complex skeleton with numerous holes. In the discussed approach, skeleton edges that do not meet the preset criteria (e.g. that the skeleton is to be located between holes in the center of the polygon), are automatically removed. A simple algorithm for smoothing zigzag lines was proposed.

It is important to reduce the computation complexity while maintaining the accuracy of convolution neural networks. We deem it is possible to further reduce the network complexity while ensuring the accuracy. In this paper, we propose a novel feature extraction network called DSRNet which is lightweight but effective. DSRNet follows the basic ideas of stacking modules and short connection, introduces Depthwise Separable convolution and utilizes the Dilated convolution. The proposed network has fewer parameters and achieves outstanding speed. We conducted comprehensive experiments on CIFAR10, CIFAR100 and STL10 datasets, and the results showed the DSRNet has great performance improvement in terms of accuracy and speed.

Network Simulators is typically used to study services and applications in complex scenarios due to the infeasibility of deploying real testbeds. Many problems can be solved by using network simulators such as NS-3. With this in mind, the aim of this article is to introduce new NS-3 users through detailed information. It is sometimes difficult to handle by new users the traditional manuals developed by NS-3 project official website. In this article, NS-3 for communication network and Eclipse Integrated Development Environment (IDE) for powerful programming language are integrated step-by-step, explaining the main features of these open source software packages and concluding with an example simulation. Our effort is to make it easy for a beginner to be part of the NS-3 research community and to maintain an open environment of knowledge.

Network Embedding (NE) is an important method to learn the representations of network via a low-dimensional space. Conventional NE models focus on capturing the structure information and semantic information of vertices while neglecting such information for edges. In this work, we propose a novel NE model named BimoNet to capture both the structure and semantic information of edges. BimoNet is composed of two parts, i.e., the bi-mode embedding part and the deep neural network part. For bi-mode embedding part, the first mode named add-mode is used to express the entity-shared features of edges and the second mode named subtract-mode is employed to represent the entity-specific features of edges. These features actually reflect the semantic information. For deep neural network part, we firstly regard the edges in a network as nodes, and the vertices as links, which will not change the overall structure of the whole network. Then we take the nodes' adjacent matrix as the input of the deep neural network as it can obtain similar representations for nodes with similar structure. Afterwards, by jointly optimizing the objective function of these two parts, BimoNet could preserve both the semantic and structure information of edges. In experiments, we evaluate BimoNet on three real-world datasets and task of relation extraction, and BimoNet is demonstrated to outperform state-of-the-art baseline models consistently and significantly.

This paper addresses the alliance route network design problem considering uncertainty of unit transportation cost. An alliance route network is constructed based on the hub-and-spoke (HS) network , in which airlines can achieve inter-area passenger transport through their international gateways. The design problem is formulated with a robust model containing a set of uncertain cost parameters. The model is established based on the three-subscript model of the HS network. A case study collected from real-world data is used to test the proposed model. The results show that the robust solution can reduce the impact of cost uncertainty.

This paper incorporates the definition of Intrusion Detection Systems and the methodologies utilised by these systems. As well as this, this research paper also encompasses a taxonomy and a survey of IDS and the specific strategies and principles. Finally, this paper also includes a discussion amongst other authors for instance what the authors differ and agree on, along with the previously related studies.

Recently, the Integrated Air Defense System (IADS) has become vital for the defense system as the military defense system is vital for national security. Placing Integrated Air Defense System batteries among locations to protect locations assets is a crucial problem because optimal solutions are needed for interceptor missiles to intercept attacker missiles for maximizing protection of assets across locations or places. In this research, the procedures of using network algorithms along with developing several network algorithms are going to be demonstrated to develop a model for sequential development of seven network securing strategies of placing Surface to Air Missile (SAM) batteries to maximize the protection of assets across locations (based on given asset values) by generating optimal solutions through computation to destroy maximum attacker missiles by using minimum interceptor missiles with given intercept probability. This network securing strategies can be implemented not only for Integrated Air Defense System (IADS) planning but also Counter Air (CA) planning as Integrated Air Defense System (IADS) is conducted with defensive counter air supported by attack operations in offensive counter air.

Convolutional Neural Networks (CNNs) are largely hand-crafted, which leads to inefficiency in the constructed network. Various other algorithms have been proposed to address this issue, but the inefficiencies resulting from human intervention have not been addressed. Our proposed EvolveNet algorithm is a task-agnostic evolutionary search algorithm that can find optimal depth and width scales automatically in an efficient way. The optimal configurations are not found using grid search, instead evolved from an existing network. This eliminates inefficiencies that emanate from hand-crafting, thus reducing the drop in accuracy. The proposed algorithm is a framework to search through a large search space of subnetworks until a suitable configuration is found. Extensive experiments on the ImageNet dataset demonstrate the superiority of the proposed method by outperforming the state-of-the-art methods.

Model understanding is critical in many domains, particularly those involved in high-stakes decisions, i.e., medicine, criminal justice, and autonomous driving. Explainable AI (XAI) methods are essential for working with black-box models such as Convolutional Neural Networks. This paper evaluates the traffic sign classifier of Deep Neural Network (DNN) from the Programmable Systems for Intelligence in Automobiles (PRYSTINE) project for explainability. The results of explanations were further used for the CNN PRYSTINE classifier vague kernels` compression. After all, the precision of the classifier was evaluated in different pruning scenarios. The proposed classifier performance methodology was realised by creating the original traffic sign and traffic light classification and explanation code. First, the status of the kernels of the network was evaluated for explainability. For this task, the post-hoc, local, meaningful perturbation-based forward explainable method was integrated into the model to evaluate each kernel status of the network. This method enabled distinguishing high and low-impact kernels in the CNN. Second, the vague kernels of the classifier of the last layer before the fully connected layer were excluded by withdrawing them from the network. Third, the network's precision was evaluated in different kernel compression levels. It is shown that by using the XAI approach for network kernel compression, the pruning of 5% of kernels leads only to a 1% loss in traffic sign and traffic light classification precision. The proposed methodology is crucial where execution time and processing capacity prevail.

Drug repositioning (also called drug repurposing) is a strategy for identifying new therapeutic targets for existing drugs. This approach is of great importance in pharmacology as it is a faster and cheaper way to develop new medical treatments. In this paper, we present, to our knowledge, the first application of multiplex-heterogeneous network embedding to drug repositioning. Network embedding learns the vector representations of nodes, opening the whole machine learning toolbox for a wide variety of applications including link prediction, node labelling or clustering. So far, the application of network embedding for drug repositioning focused on heterogeneous networks. Our approach for drug repositioning is based on multiplex-heterogeneous network embedding. Such method allows the richness and complexity of multiplex and heterogeneous networks to be projected in the same vector space. In other words, multiplex-heterogeneous networks aggregate different multi-omics data in the same network representation. We validate the approach on a task of link prediction and on a case study for SARS-CoV2 drug repositioning. Experimental results show that our approach is highly robust and effective for finding new drug-target associations.

Networks and their properties have been used to study complex biological systems. Recently, network curvature measures have demonstrated the ability to capture relevant net- work properties. This study employs network curvature measures to analyse gene expression correlations in various human tissues for identifying unique topological features that differentiate these groups. Preliminary findings suggest that curvature measures offer novel insights that could enhance our understanding of the biological systems.

This paper takes a novel approach to simulating fractional stride decreases in binary artificial neural networks (ANNs), addressing the limitations of fixed integer strides typically used in traditional convolutional neural networks (CNNs). By introducing a dynamic stride adjustment mechanism that reduces stride over time using either exponential or logarithmic decay functions, the network can simulate fractional strides, enabling the capture of increasingly fine-grained spatial details at deeper layers. This method improves output resolution, feature extraction without increasing network complexity, and possible solutions for binary ANN architectures where computational efficiency is critical. Using synthetic data, we analyze how both decay functions impact output resolution in a comparative experiment. Exponential decay produces a rapid initial stride decrease, leading to early significant resolution gains, while logarithmic decay offers a more gradual and controlled resolution increase. Both methods are enhanced by an attention mechanism that selectively emphasizes important regions in feature maps, improving spatial feature extraction efficiency. Our findings demonstrate that dynamic, asymptotic stride adjustment provides an effective method for simulating fractional strides in binary ANNs, with potential applications in areas such as object detection, segmentation, super-resolution, and autonomous systems, including spacecraft navigation and hazard detection. We also examine the trade-offs between the two decay methods in terms of resolution growth and computational overhead, contributing to ongoing research in adaptive ANN architectures and binary neural network design.

Unmanned delivery technology using unmanned aerial vehicles (UAVs) has the potential to solve issues such as traffic congestion and labor shortages. We investigated a method to construct a route between the source and destination nodes in a wireless multi-hop network by installing a radio terminal in each house. If the relay functions of all nodes in the network are in operation, the number of control messages for route construction increases, and the message relay forwarding load of each node increases. Keeping all node pairs multi-hop connectivity, we establish a method to operate the relay function with as few nodes as possible. End nodes that have not activated the relay function connect to the relay node with the closest physical distance. In this paper, we propose methods for selecting nodes to operate the relay function. Specifically, one method is based on random selection, another method is based on the number of adjacent nodes, and a third method is based on the Multipoint Relay (MPR) of the Optimized Link State Routing Protocol (OLSR). For each method, we evaluate whether the route construction is available, the number of relay nodes, the total distance of constructed routes, and the distance in population-dense regions. We clarify the characteristics of each method, and the method based on MPR is the best for constructing a network suitable for UAV navigation routes. In addition, the smaller the number of relay nodes, the more unexpected collisions on the links due to route conflicts. We clarify the minimum node density required to apply the MPR method which reduces the number of relay nodes suppressing route conflicts.

Deeper understanding of biological processes, disease mechanisms, and prospective therapeutic targets can be attained by Gene Expression Network Analysis (GENA), which offers a potent framework for revealing the intricate regulatory mechanisms controlling gene expression. GENA is a computational method used to understand the intricate interactions and relationships between the genes in a biological system. It entails using network theory, statistical analysis, and gene expression data to pinpoint functional modules, regulatory linkages, and important genes or pathways involved in a particular biological process or illness. This chapter broadly outlines the principles and practice of GENA, to an novice reader and outlines a simple method of performing a GENA using online Rice gene expression datasets available in various websites.

An earthquake is a natural event by its general definition. This natural event is a disaster that causes significant damage, loss of life, and other economic effects that will damage the state. The possibility of predicting a natural event such as an earthquake will minimize the reasons mentioned. Data collection, data processing, and data evaluation were carried out in this study. Earthquake forecasting was performed using the data and the RNN (Recurrent Neural Network) method. The study was carried out on seismic data with a magnitude of 3.0 and above belonging to Düzce Province between 1990 and 2022. In order to increase the learning potential of the method, the b and d values of the earthquake were calculated and included in the data set, except for the earthquake magnitude. The determination of earthquakes in a specific time interval in regions of Turkey, the classification of earthquake-related seismic data using artificial neural networks, and the production of predictions for the future reveal the importance of this study.

This research aims to produce a network structure in the Integrated Urban Farming Program in Bandung City to map the policy actors involved in it as an effort to support food security in the city. This research uses a mixed method with an exploratory sequential strategy involving policy actors from the government, private sector, academia, community, and mass media. To obtain the network structure in integrated urban farming to determine the most important actors, this research uses a social network analysis approach by utilizing the Gephi application. The network structure is based on four dimensions, namely centrality. The results of this study confirm that the actor who has the most connections (degree centrality), as well as the one who holds the best communication control (betweenness centrality), is Parahyangan Catholic University (Academic). While the actor who plays the most important role (eigenvector) is at the lower level of the government Sub-District and Urban Village. This research is useful for explaining the importance of the position of actors in the urban agriculture policy network which is the key to the success of a program.

The heterogeneous structure implies that a few nodes may be crucial in maintaining network structural and functional properties. Identifying these crucial nodes correctly and quickly is a primary issue as contemporarily may face the mushrooming of large-scale datasets. Besides, the ‘weight issue’ is always ignored in this field which edge weight may play a positive/negative role in contributing to the node importance in different weighted networks. This paper provides a novel algorithm, Weighted Expectation Algorithm (WEA), which aims to ensure accuracy and speed of computation by taking advantage of dynamic programming to better handle the task of large-scale networks. Additionally, the weight issue that edge weights may contribute differently is addressed by a simply quantitative definition. Two standard experiments show WEA can maintain the network structure in connectivity well (by the lowest average robustness 0.192) and identify the node importance better in the spreading function test of spreading dynamics (by the highest average Kendall’s tau-b 0.678). In addition, the time complexities of different algo-rithms are evaluated, and their time-consuming are tested, proving that WEA consumes a rela-tively short time.

The electroencephalogram (EEG) is a record of brain activity. Brain Computer Interface (BCI) technology formed by the EEG signal has become one of the hotspots at present. How to extract the feature signal of EEG is the most basic research of BCI technology. In this paper, A new method of recognizing fatigue, conscious, concentrated state of human brain is proposed by the combination of discrete wavelet transform and the neural network based on EEG signal. First of all, the law signal is preprocessed by the wavelet denoising method because the law EEG signal contains a large number of high frequency noise, which is decomposed into multi-layer high frequency signal and low frequency signal. thus, δ wave, θ wave, α wave, β wave are obtained by the wavelet transform. And then, frequency band energy of the different wave is regards as the feature signal of EEG. In the experiment, the feature signal is classified by radial basic function (RBF) and annealed chaotic competitive learning network (ACCLN). RBF and ACCLN networks are trained with 500 sets of sample data and are tested by 100 sets of samples in different mental states. The experimental results show that the average accuracy of RBF network under three conditions are 88.75%, 88.25%, 88.5%, respectively, and the correct rate of ACCLN network is 97%, 98%, 98%, respectively.

In the realm of System Biology, it is a challenging endeavor to infer a gene regulatory network from time-series gene expression data. Numerous Boolean network inference techniques have emerged for reconstructing a gene regulatory network from a time series gene expression dataset. However, most of these techniques pose scalability concerns given their capability to consider only 2 to 3 regulatory genes over a specific target gene. To overcome this limitation, a novel inference method, LBF- MI, has been proposed in this research. This two-phase method utilizes limited Boolean functions and multivariate mutual information to reconstruct a Boolean gene regulatory network from time-series gene expression data. Initially, Boolean functions are applied to determine the optimum solutions. In case of failure, multivariate mutual information is applied to obtain the optimum solutions. This research conducted a performance-comparison experiment between LBF-MI and two other methods: Context Likelihood Relatedness, and Relevance Network. When examined on artificial as well as real time-series gene expression data, the outcomes exhibited that the proposed LBF-MI method outperformed Context Likelihood Relatedness and Relevance Network on artificial datasets, and two real Escherichia Coli datasets (E. coli gene regulatory network, and SOS response of E. coli regulatory network).

Although neural network architectures are critical for their performance, how the structural characteristics of a neural network affect its performance has still not been fully explored. We here map architectures of neural network to directed acyclic graphs, and find that incoherence, a structural characteristic to measure the order of directed acyclic graphs, is a good indicator for the performance of corresponding neural networks. Therefore we propose a deep isotropic neural network architecture by folding a chain of same blocks then connecting the blocks with skip connections at different distances. Our models, named FoldNet, have two distinguishing features compared with traditional residual neural netowrks. First, the distances between block pairs connected by skip connections increase from always equal to one to specially selected different values, which lead to more incoherent graphs and let the neural network explore larger receptive fields and thus enhance its multi-scale representation ability. Second, the number of direct paths increases from one to multiple, which leads to a larger proportion of shorter paths and thus improve the direct propagation of information throughout the entire network. Image classification results on CIFAR-10 and Tiny ImageNet benchmarks suggested that our new network architecture performs better than traditional residual neural networks.

A convolutional neural network (CNN) is sometimes understood as a black box in the sense that while it can approximate any function, studying its structure will not give us any insights into the nature of the function being approximated. In other terms, the discriminative ability does not reveal much about the latent representation of a network. This research aims to establish a framework for interpreting the CNNs by profiling them in terms of interpretable visual concepts and verifying them by means of Integrated Gradient. We also ask the question, "Do different input classes have a relationship or are they unrelated?" For instance, could there be an overlapping set of highly active neurons to identify different classes? Could there be a set of neurons that are useful for one input class whereas misleading for a different one? Intuition answers these questions positively, implying the existence of a structured set of neurons inclined to a particular class. Knowing this structure has significant values; it provides a principled way for identifying redundancies across the classes. Here the interpretability profiling has been done by evaluating the correspondence between individual hidden neurons and a set of human-understandable visual semantic concepts. We also propose an integrated gradient-based class-specific relevance mapping approach that takes the spatial position of the region of interest in the input image. Our relevance score verifies the interpretability scores in terms of neurons tuned to a particular concept/class. Further, we perform network ablation and measure the performance of the network based on our approach.

Medical Trust-Network is one of the most promising fields of study in network science. Establishment of trust within medical entities ensures better treatment and increases better medical facilities. The word ‘Trust’ signifies a very important behavioral aspect between any human entities, especially among doctors and patients. To represent such relationships Trust Network Models are built to express the interactions between human entities within such networks. Though the idea of a Trust-Network has traditionally been one of the major areas of research, yet the concept of a medical trust network model is relatively a new domain. In this paper, we introduce an overall multilayered Trust Network to represent the entire healthcare architecture. More specifically our model is based on an evolutionary graph system with a discrete relationship between the three most important entities of any healthcare system, namely – Doctors, Departments, and Hospitals. Observations indicate that based on our model, the medical healthcare system is a multilayered model unlike a feed-forward model as indicated by previous studies.

With the fast development of the mobile Internet, the online platforms of social networks have rapidly been developing for the purpose of making friends, sharing information, etc. In these online platforms, users being related to each other forms social networks. Literature reviews have shown that social networks have community structure. Through the studies of community structure, the characteristics and functions of networks structure and the dynamical evolution mechanism of networks can be used for predicting user behaviours and controlling information dissemination. Therefore, this study proposes a deep community detection method which includes (1) matrix reconstruction method, (2) spatial feature extraction method and (3) community detection method. The original adjacency matrix in social network is reconstructed based on the opinion leader and nearer neighbors for obtaining spatial proximity matrix. The spatial eigenvector of reconstructed adjacency matrix can be extracted by an auto-encoder based on convolution neural network for the improvement of modularity. In experiments, four open datasets of practical social networks were selected to evaluate the proposed method, and the experimental results show that the proposed deep community detection method obtained higher modularity than other methods. Therefore, the proposed deep community detection method can effectively detect high quality communities in social networks.

In this perspective article, we ponder topographical enhancements of artificial neural networks. In our recent paper in JMLR, we proved a quasi-equivalence between the network width and depth and also discussed the power of intra-links, which can be viewed as network height. In 1982, Hopfield published a network to model human associative memory, which contains many loops for dynamic evolution toward fixed points. Based on noising-denoising loops, diffusion models are recently developed to enable Bayesian modeling and inference with big data. Furthermore, we envision development of multi-AI-agent systems through “netware” engineering as a quantum leap of software engineering for emergent behaviors and autonomous AI at individual and population levels. We believe that the novel use of links and loops in space and time via multi-scale coupling would catalyze the next-generation neural networks.

Our objective in this work is to demonstrate how Physics-Informed Neural Networks, a type of deep learning technology, can be utilized to examine the mechanical properties of a helicopter blade. The blade is regarded as a prismatic cantilever beam that is exposed to triangular loading, and comprehending its mechanical behavior is of utmost importance in the aerospace field. PINNs utilize the physical information, including differential equations and boundary conditions, within the loss function of the neural network to approximate the solution. Our approach determines the overall loss by aggregating the losses from the differential equation, boundary conditions, and data. We employed a Physics-Informed Neural Network (PINN) and an Artificial Neural Network (ANN) with equivalent hyperparameters to solve a fourth-order differential equation. By comparing the performance of the PINN model against the analytical solution of the equation and the results obtained from the ANN model, we have conclusively shown that the PINN model exhibits superior accuracy, robustness, and computational efficiency when addressing high-order differential equations that govern physics-based problems. In conclusion, the study demonstrates that PINN offers a superior alternative for addressing solid mechanics problems with applications in the aerospace industry.

Phylostratigraphic analysis is a way to look anew on phylogenetic data in the evolutionary aspect. It allows counting the evolutionary age based on the analysis of genes, their orthologs and finding the last common ancestor. We performed phylostratigraphic analysis of Arabidopsis thaliana genes associated with several types of abiotic stresses (heat, cold, water-related, light, osmotic, salt, and oxidative) determined by the Gene Ontology annotation. Comparison of the distributions of ages of genes associated with stresses of different type has shown the heat stress to involve older genes while the light stress – younger genes. At the same time, all types of stress are characterized by a significantly higher proportion of old genes (common to all eukaryotes) compared to the whole set of A.thaliana genes. This can be explained by the involvement of basic molecular processes in plant cells into the stress response. Reconstruction and graphical analysis of the gene network of the heat stress educed several clusters associated with different response functions. Some of these clusters contain only ancient genes. The results obtained show that the phylostratigraphic analysis reveals the fundamental features of the organization of gene networks and their evolution.

Dyslipidemia (Dys) is a disease characterised by abnormally high levels of lipids in the 1 blood. It causes the deposits of lipids on the arterial wall, leading to the insurgence of atherosclerosis, 2 i.e. the increased thickness of arterial walls, which finally increases the risk of cardiovascular diseases. 3 Dyslipidemia is as comorbidity of Type 2 Diabetes Mellitus (T2DM). A set of genes has been associated 4 with dyslipidemia, but examining the regulation of such genes in age and sex is still an open research 5 field. In this study, starting from publicly available databases, we select genes associated with 6 dyslipidemia and we analyse their basal level changes by means of age and sex. Also studying the 7 networks associated to the genes, we identify the rewiring changes showing that there exists a set of 8 genes whose expression changes with age in: artery tibial, artery aorta and blood. Finally, we report 9 about changes associated to gender atributes in the extracted data

Dyslipidemia (Dys) is a disease characterised by abnormally high levels of lipids in the blood. It causes the deposits of lipids on the arterial wall leading to the insurgence of atherosclerosis, i.e. the increased thickness of arterial walls, which finally increases the risk of cardiovascular diseases. Dyslipidemia is a comorbidity of Type 2 Diabetes Mellitus (T2DM). A set of genes has been associated with dyslipidemia, but examining the regulation of such genes in age and sex is still an open research field. In this study, we select from publicly available databases the genes associated with dyslipidemia and analyse changes in the basal level of such genes by age and sex. Moreover, we also distinct differential rewiring of the related gene association network in males and females. Our results show that there exist a set of genes whose expression changes with age in artery tibial, artery aorta and blood. Moreover, there exist significant differences between sex.

Protein Contact Network (PCN) is an emerging paradigm for modelling protein structure. A common approach to interpreting such data is through network-based analyses. It has been shown that clustering analysis may discover allostery in PCN. Nevertheless Network Embedding has shown good performances in discovering hidden communities and structures in network. In this work, we compare some approaches for graph embedding with respect to some classical clustering approaches for annotating protein structures.

Previous high throughput screening studies indicated trans-eQTLs tend to be tissue specific. In this study, I probed if this feature can be used to identify tissue-specific gene regulatory networks. eQTL data (P<5e-8) for 16,259 genes were identified and their summary association statistics were obtained from the eQTLGen database. Next, eQTLs that display both cis and trans regulatory effects were selected and the association between their corresponding genes were examined by Mendelian randomization. 253 genes that exerted trans-regulatory impacts on 1,439 genes were identified. Majority of the identified gene-pairs aggregated into a gene regulatory network significantly enriched in hemo-immune processes. The robustness of finding was confirmed through simulation. Genes with higher gene expression heritabilities displayed broader regulatory impact in the network. The network displayed the scale-free topology, this provided the reason to examine the association of the network’s hub genes with the phenome. The outcome of analyses revealed GSDMB and ORMDL3 impact several disorders of immune origin and higher expression of HHEX is associated with lower triglycerides and risk of type 2 diabetes. This study reports trans-eQTLs can be used to identify tissue-specific gene regulatory networks and describes a workflow to achieve this purpose. The identified network displayed scale-free topology indicating the hub genes of a tissue-specific network could be targeted to prevent disease outcomes.

Several effective online and offline inference tools offer an interactive platform for inference, visualization, and analysis of gene regulatory networks. However, a tool currently needs to offer a scalable, lightweight, integrated online platform for carrying out inference, visualization, benchmarking, and extensive network analysis in an integrated manner. We introduce NetRA (Network Reconstruction and Analysis for Gene Regulatory Network), a comprehensive web tool for network analysis, visualization, and inference of large expression networks. Additionally, a platform for benchmarking and evaluation is provided. In order to deliver a highly scalable, lightweight, and rich user interface, the tool is created using the most recent technologies. We incorporate the original code of eleven (11) inference algorithms (not limited to) from Bioconductor in the current version. We also report a qualitative comparison of 11 candidate methods regarding prediction accuracy and network characteristics which will shed light on their performance against 4000 sizes synthetic DREAM network. We believe the tool will significantly aid biologists' downstream system biology research. NetRA tool can be extensible to different kinds of complex network analyses. Availability: On request (sroy01@cus.ac.in).

Keywords: Convolutional Neural Network, Deep Learning, Skin cancer, Generative Adversarial Network

Scientific social networking sites like Researchgate or Academia.edu have become part of the work practice of academic researchers. These digital platforms have been designed precisely to encourage the exchange of knowledge between scholars and to help the expansion of collaborative networks among them. Even if studies on this topic have multiplied in recent years, there is a dearth of research on the actual impact of these platforms on scientific production. The goal of this study is to investigate the relation between the attitude of researchers towards scientific social networks and the use of knowledge in their scientific work. Data from users of the scientific social networking site Researchgate were collected. A total of 143 valid responses were received and structural equation modeling was applied for data analysis. Findings of this study confirmed that researchers use knowledge obtained from scientific social networking sites both incorporating it within their research products and to acquire new competences. In particular approaching the platform as a scientific community with a shared language and a shared vision was found to have a positive impact on knowledge use. To the best knowledge of the authors this is the first study investigating the actual use of knowledge from scientific social networking sites by academic researchers. Results help to understand the impact of these platforms on the work practice of a strategic sector like scientific research.

Animal-focused research has been crucial for scientific advancement however, in this matter, rodents are still taking a starring role. Coming out from merely being supportive of evidence found in rodents, the use of fish models has slowly taken a more central role and expanded its overall contributions in areas such as social sciences, evolution, physiology, and recently in translational medical research. In neurosciences, zebrafish has been widely adopted, contributing to our understanding of the genetic control of brain processes, and the effects of pharmacological manipulations. However, discussion continues regarding the paradox of function versus structure, when fish and mammals are compared, and on the potentially evolutionarily conserved nature of behaviour across fish species. From the behavioural stand point we explored aversive/stress and social behaviour in selected fish models, and refer to the extensive contributions of stress and monoaminergic systems. We suggest that, in spite of marked neuroanatomical differences between fish and mammals, stress and sociality are conserved at the behavioural and molecular levels. We also suggest that stress and sociality are mediated by monoamines in predictable and non-trivial ways, and that monoamines could “bridge” the relationship between stress and social behaviour. To reconcile the level of divergence with the level of similarity, we need neuroanatomical, pharmacological, behavioural, and ecological studies conducted in the laboratory and in nature. These areas need to add to each other to enhance our understanding of fish behaviour and ultimately how this all may translate to better model systems for translational studies.

Speech Emotion Recognition (SER) is an interesting and difficult problem to handle. In this paper, we deal with it through the implementation of deep learning networks. We have designed and implemented six different deep learning networks, a Deep Belief Network (DBN), a simple deep neural network (SDNN), a LSTM network (LSTM), a LSTM network with the addition of an attention mechanism (LSTM-ATN), a Convolutional neural network (CNN), and a Convolutional neural network with the addition of an attention mechanism (CNN-ATN), having in mind, apart from solving the SER problem, to test the impact of attention mechanism to the results. Dropout and Batch Normalization techniques are also used to improve the generalization ability (prevention of overfitting) of the models as well as to speed up the training process. The Surrey Audio-Visual Expressed Emotion database (SAVEE), and the Ryerson Audio-Visual Database (RAVDESS) database were used for training and evaluation of our models. The results showed that networks with the addition of the attention mechanism did better than the others. Furthermore, they showed that CNN-ATN was the best among tested networks, achieving an accuracy of 74% for the SAVEE and 77% for the RAVDESS dataset, and exceeded existing state-of-the-art systems for the same datasets.

Speech recognition, or automatic speech recognition (ASR), is a technology designed to convert spoken language into text using software. However, conventional ASR methods involve several distinct components, including language, acoustic, and pronunciation models with dictionaries. This modular approach can be time-consuming and may influence performance. In this study, we propose a method that streamlines the speech recognition process by incorporating a unified recurrent neural network (RNN) architecture. Our architecture integrates a convolutional neural network (CNN) with an RNN and employs a connectionist temporal classification (CTC) loss function. Key experiments were carried out using a dataset comprising 576,656 valid sentences, using erosion techniques. Evaluation of the model performance, measured by the word error rate (WER) metric, demonstrated remarkable results, achieving a WER of 2%. This approach has significant implications for the realm of speech recognition, as it alleviates the need for labor-intensive dictionary creation, enhancing the efficiency and accuracy of ASR systems, and making them more applicable to real-world scenarios. For future enhancements, we recommend the inclusion of dialectal and spontaneous data in the dataset to broaden the model's adaptability. Additionally, fine-tuning the model for specific tasks can optimize its performance for targeted objectives or domains, further enhancing its effectiveness in those areas.

Financial fraud refers to the act of obtaining financial benefits through dishonest means. Such acts not only disrupt the order of the financial market, but also harm social and economic development, and breed other illegal and criminal acts. With the proliferation of the internet and online payment methods, many fraudulent activities and money laundering have shifted from offline to online, posing a significant challenge for regulators. In this work, we proposed a novel detection model by utilizing the graph neural network. Specifically, the general model structure includes the combination of Graph Convolutional Network (GCN) and Graph Attention Network (GAT), which can effectively capture the complex relationships and features in the financial transaction network. By building a multilayer graph neural network, the model can perform deep learning on the implicit patterns between nodes, thereby improving the accuracy of fraud detection. Experimental analysis results show that the performance of the proposed method on multiple public datasets is better than that of traditional methods, showing its potential in practical application.

The Chinese stock market, marked by rapid growth and significant volatility, presents unique challenges for investors and analysts. A-share stocks, traded on the Shanghai and Shenzhen exchanges, are crucial to China’s financial system and offer opportunities for both domestic and international investors. Accurate stock recommendation tools are vital for informed decision-making, especially given the ongoing regulatory changes and economic reforms in China. Current stock recommendation methods often fall short, as they typically fail to capture the complex inter-company relationships and rely heavily on financial reports, neglecting the potential of unlabeled data and historical price trends. In response, we propose a novel approach that combines graph-based structures with historical price data to develop self-learned stock embeddings for A-share recommendations. Our method leverages self-supervised learning, bypassing the need for human-generated labels and autonomously uncovering latent relationships and patterns within the data. This dual-input strategy enhances the understanding of market dynamics, leading to more accurate stock predictions. Our contributions include a novel framework for label-free stock recommendations with modeling stock connections and pricing information, and empirical evidence demonstrating the robustness and adaptability of our approach in the volatile Chinese stock market.

This study examined the tourism spatial distribution of nine cities in the Fujian province and assessed the impacts of COVID-19. The modified gravity model found that it was widely dispersed, with uneven and relatively independent tourism development in different cities. The social network analysis showed that tourism connections across cities were significantly reduced after the pandemic. The impacts of brand awareness and transport accessibility on spatial networks were positive in pre-pandemic but became negative during the pandemic. In contrast, tourist volume had negative impacts on spatial networks in pre-pandemic but had positive ones during the pandemic. Tourism resources and market performance had significantly positive impacts in the post-pandemic era. These findings provide advice on tourism recovery and destination management in coping with future critical events. It also contributes to the theoretical gravity framework in tourism and the research scope of the social networks analysis at the city level.

Computer-aided image diagnostics (CAD) have been used in many fields of diagnostic medicine. It relies heavily on classical computer vision and artificial intelligence. Quantum neural network (QNN) has been introduced by many researchers around the world and presented recently by research corporations such as Microsoft, Google, and IBM. In this paper, the investigation of the validity of using the QNN algorithm for machine-based breast cancer detection was performed. To validate the learnability of the QNN, a series of learnability tests were performed alongside with classical convolutional neural network (CCNN). QNN is built using the Cirq library to perform the assimilation of quantum computation on classical computers. Series of investigations were performed to study the learnability characteristics of QNN and CCNN under the same computational conditions. The comparison was performed for real Mammogram data sets. The investigations showed success in terms of recognizing the data and training. Our work shows better performance of QNN in terms of successfully training and producing a valid model for smaller data set compared to CCNN.

Recently, with the rapid advancement of the 4th Industrial Revolution, there has been a continuous outlook that Data (D), Network (N), and Artificial Intelligence (A) should be heavily utilized as resources in light of new value. South Korea is preparing policy training plans the likes of 5G network enhancement, AI manpower training, and data openings, etc., recognizing DNA technology as the key driving force behind the 4th Industrial Revolution and 5G network enhancement, AI manpower training, and data opening, etc. However, there is a criticism that the DNA industry remains in an early stage. Despite policy support within South Korea, the DNA industry has not been revitalized due to various causes, and in particular, not being concentrated on limited resources selectively is also pointed to as an important cause. this study would quantitatively draw the priority of policy measures as well as propose policy implications for the revitalization of the DNA industry by means of the ISM-ANP model. It is judged that South Korea’s experience in this way will be an important lesson for other countries which would politically promote the DNA industry as a new growth engine within the 4th Industrial Revolution environment.

Wireless Sensor Networks (WSNs) is considered to be a promising research area since it entails sensing capabilities. WSNs are confronted by fabricated explosion injection attacks and in return results into draining of vitality on sensor nodes routing mainly because of increased capabilities in a broad range of applications where sensor nodes use a battery as the power source. In this paper, we reveal and claim a state-of-the-art model to minimize vitality depletion of nodes within the network in return increases network period to more realistic stretch. It is revealed that the main cause is vitality consumption unbalances in different parts of the network direct resulting from having a static sink. More still, the network nodes neighboring the sink insistently partake in progressing data of other nodes to the sink. Consequently, the network nodes are supplementarily subjected to untimely vitality exhaustion. This paper mainly provides a novel movable (mobile) fuzzy logic sink approach and an intelligent system like expect systems that are able to scrutinize network situation, and make the superlative choice for spreading the network period robotically.

In this study, a computational method is proposed that employs Recurrent Convolutional 1 Neural Networks, utilizing meteorological radar images to forecast storm movement and intensity up 2 to 3 hours ahead, a process known as nowcasting. For this purpose, images from a radar situated in 3 southern Brazil were used. These data are publicly accessible on the website of the National Institute 4 for Space Research (INPE) in Brazil. The approach involves evaluating a spatiotemporal learning 5 recurrent convolutional neural network called PredRNN++. The results were validated through 6 case studies of storms within the radar’s coverage area. To evaluate the performance of the neural 7 network, both visual assessments and metrics such as RMSE and SSIM were employed. The findings 8 indicate that PredRNN++ was effective in simulating the shape and location of the meteorological 9 system.

With the emergence of Advance Logistics, the logistics industry has witnessed significant advancements, leading to a greater reliance on intelligent technologies. These technologies play a crucial role in gathering and transmitting logistics data, but they also introduce security and privacy risks to logistics management systems. The sharing of customer-sensitive information among stakeholders for efficient operations becomes vulnerable to unauthorized access, thus compromising privacy. This, amongst others, is a critical problem that needs to be addressed. The unique features of a cryptographic hash function, including immutability, efficient verification, and anonymity, make it a transformative technology that has the potential to address the mentioned challenges in various sectors. This disruptive technology offers solutions by ensuring data integrity, enabling quick verification, and preserving privacy. The findings suggest that employing Cryptographic Hash Functions to share customer data among logistics partners offers robust protection against cyber-attacks. The proposed system guarantees asset security, and as demonstrated by the case study evaluation, exhibits strong resilience in safeguarding against various security and privacy threats. Although this study is significant, it does have limitations. One of these limitations is the considerable overlap between logistics research and supply chain management (SCM). While our focus was on logistics tasks, we did not extensively delve into the broader context of SCM

Optical artificial neural networks (ONNs) — analog computing hardware tailored for machine learning [1,2] — have significant potential for ultra-high computing speed and energy efficiency [3]. We propose a new approach to architectures for ONNs based on integrated Kerr micro-comb sources [4] that is programmable, highly scalable and capable of reaching ultra-high speeds. We experimentally demonstrate the building block of the ONN — a single neuron perceptron — by mapping synapses onto 49 wavelengths of a micro-comb to achieve a high single-unit throughput of 11.9 Giga-FLOPS at 8 bits per FLOP, corresponding to 95.2 Gbps. We test the perceptron on simple standard benchmark datasets — handwritten-digit recognition and cancer-cell detection — achieving over 90% and 85% accuracy, respectively. This performance is a direct result of the record small wavelength spacing (49GHz) for a coherent integrated microcomb source, which results in an unprecedented number of wavelengths for neuromorphic optics. Finally, we propose an approach to scaling the perceptron to a deep learning network using the same single micro-comb device and standard off-the-shelf telecommunications technology, for high-throughput operation involving full matrix multiplication for applications such as real-time massive data processing for unmanned vehicle and aircraft tracking.

This study addresses the challenge of predicting the dynamic behavior of the structures under seismic excitation. Accurate prediction of such systems' responses is critical for the design and evaluation of buildings and infrastructure. Traditional methods, including numerical models and differential equation solvers, often face significant computational burdens, especially with nonlinear hysteretic behaviors and large-scale problems. To overcome these limitations, a novel Fuzzy-based Convolutional Neural Network (FuzzyCNN) model is developed. This model integrates fuzzy logic principles with convolutional neural networks to effectively manage the uncertainties and complexities inherent in soil-structure interaction under seismic loads. The model's performance is validated through both numerical simulations and experimental data from a mid-rise concrete building subjected to seismic events. Comparative analysis with a traditional Physics-informed CNN (PhyCNN) model demonstrates the superior accuracy and robustness of the FuzzyCNN in predicting seismic responses. Key results show that the FuzzyCNN model not only enhances prediction accuracy but also handles uncertainties more effectively than the PhyCNN model. The findings suggest that the FuzzyCNN model can significantly improve the efficiency and accuracy of dynamic response predictions. This advancement offers valuable implications for engineering design, seismic risk assessment, and the development of more resilient infrastructure.

The advent of the era of big data has led to the emergence of heterogeneous network data fusion as a prominent area of research. Heterogeneous network data is characterised by multi-modality, multi-source, and high dimensionality, which presents significant challenges for traditional data fusion methods. These methods often encounter difficulties in processing such data, including issues such as information redundancy, data inconsistency, and high computational complexity. This paper proposes a heterogeneous network data fusion model based on a deep neural network. The model employs the Multi-Layer Perceptron (MLP) as its fundamental framework, utilising the deep neural network to facilitate joint feature representation learning on data from disparate modalities. The Adaptive Feature Reconstruction Module enables the model to learn the interrelationships between different modalities and to balance the importance of different modal features in the fusion process in a dynamic manner. Furthermore, we introduce an innovative cross-modal attention mechanism, which is capable of effectively capturing the coupling relationship between deep features in heterogeneous data, thereby enhancing the expressiveness and data fusion efficacy of the model. The experimental results demonstrate that the proposed model markedly enhances the accuracy of classification and regression tasks in comparison to traditional methodologies.

Recently, deep neural networks (DNNs) have been widely used in various fields such as autonomous vehicles and smart homes. Since these DNNs can be directly implemented on edge devices, they offer advantages such as real-time processing in low-power and low-bandwidth environments. However, deployment of DNNs in embedded systems, including edge devices, exposes them to threats such as fault injection attacks. In this paper, we introduce two methods to induce misclassifications by using clock and voltage glitch-based fault attacks in devices where DNN models are executed. As a result of experiments on a microcontroller implemented with a DNN for image classification, we show that glitch injection attacks can lead with high probability to serious misclassifications. Furthermore, we propose a countermeasure to glitch attacks on the softmax function, and confirm that this method is effective in preventing misclassifications.

Aim: To elucidate the international network structures and evolution of collaborations in capsule endoscopy research from 2000 to 2023 by analyzing co-authorship networks. Method: This study analyzed co-authorship patterns in capsule endoscopy research articles published from 2000 to 2023 using data from the Web of Science (WoS). I employed Python (version 3.10.5) in the PyCharm development environment (version 2022.1.3) to conduct a network analysis using social network analysis principles. The analysis included macro-level metrics (network density, clustering coefficient, components, and average path length) and micro-level metrics (degree centrality, closeness centrality, and betweenness centrality) to assess collaboration intensity, key contributors, and network cohesion. Result: The co-authorship network analysis revealed evolving collaboration patterns over three distinct periods (2000-2009, 2010-2019, and 2020-2023). Initially, the network exhibited low density and high fragmentation, with key contributors such as Guido Costamagna and Jonathan A. Leighton prominently positioned. Over time, the network expanded with increased density and clustering, though it remained fragmented. In recent years, the network showed increased cohesion with reduced components and notable contributors like Anastasios Koulaouzidis and Xavier Dray, reflecting the growing importance of international collaborations, particularly involving researchers from Japan, Italy, and Israel. Conclusion: The study highlights the dynamic nature of international collaborations in capsule endoscopy research, with increasing connectivity and influential contributors playing pivotal roles in advancing the field. These findings underscore the value of international cooperation in enhancing diagnostic and therapeutic approaches in capsule endoscopy, pointing to future opportunities for deeper collaboration and integration.

This paper investigates the use of Physics-Informed Neural Networks (PINNs) in active learning cycles. We defined two scenarios: one semi-supervised and the other fully supervised. PINNs emphasize the integration of physical laws into neural networks to improve the predictive performance of vanilla neural networks and to enhance the efficiency of traditional methods for solving partial differential equations (PDEs). Key contributions include adapting existing computational frameworks to enable the use of Graph Neural Networks for solving problems that require the calculation of gradients on unstructured triangle meshes, a query strategy focusing on the physical loss, and a comparative analysis of this strategy against random sampling across both defined scenarios. This work establishes a foundation for future research aimed at expanding the application of Physics-Informed Graph Neural Networks (PIGNN) using active learning and addressing real-world problems in fluid dynamics and electrodynamics.

The application of artificial intelligence to point-of-care testing (POCT) disease detection has become a hot research field, in which breath detection which detects the patient's exhaled VOCs combined with sensor arrays of Convolutional Neural Network (CNN) algorithms as a new lung cancer detection is attracting more researchers' attention. However, due to low accuracy, high complexity computation and large number of parameters make the CNN algorithms difficult transplant to the embedded system of POCT devices. A lightweight neural network (LTNet) in this work is proposed to deal with this problem, and meanwhile, achieve high-precision classification of acetone and ethanol gases, which are respiratory markers for lung cancer patients. Compared to currently popular lightweight CNN models such as EfficientNet, LTNet has fewer parameters (32K) and its training weight size is only 0.155 MB. LTNet achieved an overall classification accuracy of 99.06% and 99.14% in the own mixed gas dataset and the University of California (UCI) dataset, which are both higher than that of the five existing models’, and it also offers the shortest training (844.38s and 584.67s) and inference time (23s and 14s) in the same validation set. Compared to the existing CNN models, LTNet is more suitable for resource-limited POCT devices.

Next-generation cancer and oncology research needs to take full advantage of the multi-modal structured, or graph, information, with the graph datatypes ranging from molecular structures to spatially resolved imaging and digital pathology to biological networks to knowledge graphs. Graph Neural Networks (GNNs) efficiently combine the graph structure representations with the high predictive performance of deep learning, especially on the large multi-modal datasets. In this review article, we survey the landscape of recent (2020-present) GNN applications in the context of cancer and oncology research, and delineate six currently predominant research areas. Subsequently, we identify the most promising directions for future research. We compare GNNs with graphical models and "non-structured" deep learning, and devise the guidelines for cancer and oncology researchers or physician-scientists asking the question of whether they should adopt the GNN methodology in their research pipelines.

Relay protection service is the most crucial service carried by the electric power communication network, which is directly linked to the safe operation of the power grid. In the route planning of power services, a route planning architecture for the power communication network is constructed in this paper, considering the operation and maintenance requirements, such as relay protection service overload (RPSO) and dual route. To minimize network risk variance, the problem of single-service route allocation is formulated by taking the factors into account such as service delay, link bandwidth utilization, and RPSO constraints. Based on $k$ shortest paths algorithm, a route planning scheme for the power services is proposed, which considers RPSO and risk balance, and a strategy for implementing multi-service route planning of the communication network is studied. The simulation results show that the proposed route planning strategy can effectively avoid relay protection service overload, reduce network risk variance, and balance network risk.

In the trend of transforming existing systems and services into non-face-to-face models, the concert industry is also showing movements toward transitioning to virtual formats. Physical concerts in the real world require venues that can accommodate hundreds to tens of thousands of spectators, but non-face-to-face methods that can accommodate large audiences face various limitations. Moreover, to elevate the satisfaction level of virtual concert attendees to that of real-world concerts, it's important to implement interaction between performers and audiences. Modern metaverse platforms apply cutting-edge network technologies to accommodate numerous users within a single channel. Many researchers are adopting network technologies such as SDN (Software Defined Networking) and CDN (Content Delivery Network) to set up a virtual concert that can accommodate large audiences. In this paper, we propose a network framework to be designed for the composition of virtual concerts. In particular, we separate a channel dedicated to interaction in order to provide an immersive experience of exchanging interactions between performers and audiences. As massive audiences transmitting interaction data to the performer in a 1:N format can lead to problems with acceptance and latency, this study introduces a concept of a channel form called 'Zone' and proposes an interaction data channel network framework that does not compromise immersion. The proposed framework supports tasks for effectively transmitting interaction data using network technologies for metaverse platforms like XDN and clustering algorithms like fuzzy c-means. We also suggest a CDN-based architecture that can ensure low latency for performers to transmit interaction data to the audience.

Coastal marine ecosystems worldwide, like Florida Bay, are increasingly affected by tide alteration and anthropogenic disturbances which affect water quality imbalance leading to algal blooms. Increased bloom persistence is a serious threat due to the long-lasting impact on processes like carbon cycle and services like species presence. However, exploring eco-environmental feedback patterns of algal blooms remains challenging and poorly investigated, also due to the paucity of data. Florida Bay, taken as an epitome, has long experienced algal blooms in its central and western regions, and in 2006 an unprecedented bloom occurred in the eastern habitats. We analyzed the occurrence of blooms from three perspectives: (1) the spatial spreading networks of chlorophyll-a (CHLa) that pinpoints source and unbalanced habitats; (2) the fluctuations of water quality factors pre- and post-bloom outbreaks to in assess environmental impacts of ecological dysbiosis and target prevention and control of algal of blooms; and (3) the biogeochemical-spreading network topological co-evolution to quantify ecosystemic stability and ecological shift likelihood in long-term. Here, we propose Transfer Entropy (TE) to quantify dynamical interactions between spatial areas and biogeochemical factors (ecosystem connectome) underpinning bloom emergence and spreading, as well as environmental effects; the Pareto principle is defined for identifying the salient eco-environmental interactions of CHLa. We quantified the spatial dynamics of algal blooms, and thus obtained areas in need for ecological monitoring and potential bloom control. Results show that algal blooms are increasingly persistent over space with long-term negative effects on water quality factor. A dichotomy is reported between spatial ecological corridors of spreading and biogeochemical networks as well as divergence from the optimal eco-organization: randomization of the former due to nutrients' overload and temperature increase leads to scale-free CHLa spreading and extreme outbreaks a posteriori. Subsequently this increases bloom persistence, turbidity and salinity with potentially strong ecological effects on highly biodiverse and vulnerable areas such as tidal flats and marshes. Algal blooms are important ecosystem regulators of nutrient cycles; however, beyond limit chlorophyll-a outbreaks cause aquatic species mortality due to their effect on water turbidity, nutrient balance (nitrogen and phosphorus in particular), salinity and temperature. Beyond compromising local environmental quality, socio-ecological services are also compromised at large scales, yet ecological assessment models like the one presented are in need of application in subtropical and tropical bays worldwide.

With the development of the city, a large number of water networks in the built-up areas of Bangkok have been filled and hardened, resulting in poor urban flooding and aggravating flooding, causing loss of life and property of citizens. In this paper, on the basis of combing the current water networks and open space potential flood storage points in the built-up areas of Bangkok, the complex network diagram of the water system in the built-up areas of Bangkok is constructed by combining the theory of complex networks and analyzing the attribute parameters of the network and the characteristic parameters of the open space storage nodes and water system paths, and finding that the water system network in the built-up area of Bangkok has complex network characteristics such as robustness, clustering and hierarchy. By exploring the key storage points and water system connection paths, the researchers initially constructed a flood safety pattern in the built-up area of Bangkok with 145 key nodes and 127 river paths as the backbone, and conceptualized the development study of the flood safety pattern in both horizontal and vertical directions. The urban flood safety pattern based on complex network theory proposed in this paper provides a case reference and methodological ideas to scientifically solve the game conflict between the demand for construction land for urban development and the construction area of urban open space storage points and water storage network under the increasingly severe flooding situation.

Forest fire identification is important for forest resource protection. Effective monitoring of forest fires requires the deployment of multiple monitors with different viewpoints, while most traditional recognition models can only recognize images from a single source. By ignoring the information from images with different viewpoints, these models produce high rates of missed and false alarms. In this paper, we propose a graph neural network model based on the similarity of dynamic features of multi-view images to improve the accuracy of forest fire recognition. The input features of the nodes on the graph are converted into relational features of different gallery pairs by establishing pairs (nodes) representing different viewpoint images and gallery images. The new feature library relationship is used to update the image gallery with dynamic features in order to achieve the estimation of similarity between images and improve the image recognition rate of the model. In addition, to reduce the complexity of image pre-processing process and extract key features in images effectively, this paper also proposes a dynamic feature extraction method for fire regions based on image segment ability. By setting the threshold value of HSV color space, the fire region is segmented from the image, and the dynamic features of successive frames of the fire region are extracted. The experimental results show that, compared with the baseline method Resnet, this paper's method is more effective in identifying forest fires, and its recognition accuracy is improved by 2%. And the scheme of this paper can adapt to different forest fire scenes, with better generalization ability and anti-interference ability.

Over the past decade, wireless communication technologies have developed significantly for intelligent applications in road transport. This paper provides an overview of telecommunications-based intelligent transport systems with a focus on ensuring system safety and resilience. In vehicle-to-everything, these problems are extremely acute due to the specifics of the operation of transport networks, which requires the use of special protection mechanisms. In this regard, it was decided to use blockchain as a system platform to support the needs of transport systems for secure information exchange. This paper describes the technological aspects of implementing blockchain technology in vehicle-to-network; the features of such technology are presented, as well as the features of their interaction.

One of the most serious concerns about mega event-related changes to small cities is how to effectively utilize newly developed public attractions after the Olympic Games. Making connections with existing local amenities and forming attraction networks can be effective strategies for continuing a city’s revitalization. However, despite the expected benefits, attraction network research shows that these benefits often fail to materialize. With the case of Gangneung, a 2018 winter Olympic hosting city, this study investigated visitation patterns to nineteen selected attractions using network analysis. The results indicate that the most influential nodes are located on the northern coast, the eastern coast, and in the south downtown area, those nodes being the central locations where the strongest of connections are made. New attractions such as the Olympic Park and Walwha Linear Park were rather isolated. While seasonal and periodic variations, visitors’ residences, and destination choice attitudes had a significant effect on visitation patterns, the attraction networks, modes of transportation, proximity to region, and type similarity were not significant factors in the forming of visitation patterns. The results make a methodological contribution to tourist behavior and network research. In addition, beyond individual attraction development, the results provide practical implications in regard to networking and cooperation between multiple attractions using temporal and spatial strategies such as management/investment prioritizing, travel route development, and program scheduling.

Alzheimer’s Disease (AD) is a neurodegenerative condition that currently has no known cure. The principles of the expanding field of network medicine (NM) have recently been applied to AD research. The main principle of NM proposes that diseases are much more complicated than one mutation in one gene, and incorporate different genes, connections between genes, and pathways that may include multiple diseases to create full scale disease networks. AD research findings as a result of the application of NM principles have suggested that functional network connectivity, myelination, myeloid cells, and genes and pathways may play an integral role in AD progression, and may be integral to the search for a cure. Different aspects of the AD pathology could be potential targets for drug therapy to slow down or stop the disease from advancing, but more research is needed to reach definitive conclusions. Additionally, the holistic approaches of network pharmacology in traditional Chinese medicine (TCM) research may be viable options for the AD treatment, and may lead to an effective cure for AD in the future.

Linear state estimation (SE) formulation under a rectangular coordinate system has been proved to be applicable for real-time distribution network management. Micro phasor measurements’ model can be accommodated into this kind of SE easily. However, voltage magnitude, active power and reactive power measurements are transformed to linear measurements with large node voltage error. To cope with this issue, a linear state estimation under a polar coordinate system is adopted at the first stage to obtain accurate enough complex node voltage, and then nonlinear measurements are transformed to be linear with complex node voltage. At the second stage, linear SE under a rectangular coordinate system can be adopted to satisfy more strict network constraints. The proposed two-stage linear SE is validated on balanced 14, 33, 70,84, 119, 135 nodes network and IEEE 13, 34, 37,123 unbalanced test feeders.

Stock price prediction is of great significance but faces numerous challenges. In this study, a hybrid CNN-LSTM model was utilized. The stock data were processed and feature engineering was conducted. The model architecture and training strategies were expounded. Experimental results demonstrated that this model outperformed traditional methods and benchmark models on the test set, featuring strong capabilities in feature extraction and handling long-term dependencies. Key roles of data preprocessing, hyperparameter adjustment, and model fusion were also summarized, providing valuable references for stock price prediction.

This study investigates the integration of natural intelligence (NI) and artificial intelligence (AI) within traditional firms, proposing a multilayer network framework to enhance economic value. The research explores the central question: How can the integration of NI and AI, facilitated by copula nodes, drive economic value creation in digitized firms? Through empirical testing across multiple industries—namely manufacturing, retail, and finance—the study evaluates the cost-benefit improvements associated with AI adoption. The findings reveal substantial efficiency gains, improved decision-making, and notable cost reductions, validating the proposed model's predictions. The study emphasizes the critical role of copula nodes in optimizing interactions between NI and AI, ensuring that their integration yields maximum economic benefits. By providing a novel framework for quantifying AI's economic impact within digitized firms, this research fills a significant gap in the literature on multilayer networks and intelligent systems. The practical implications are profound, offering a strategic pathway for firms to enhance profitability and competitiveness in the digital age.

The causal networks are important for understanding the signaling alterations in diseases. To reveal the network pathways affected in epithelial-mesenchymal transition (EMT) and cancer stem cells (CSCs), which are related to poor prognosis of cancer, the molecular networks and gene expression in diffuse- and intestinal-type gastric cancer (GC) have been analyzed. The network pathways in GC were analyzed in Ingenuity Pathway Analysis (IPA). The analysis of the probe sets of which gene expression has significant differences between diffuse- and intestinal-type GC in RNA Sequencing publicly available data identified 1099 causal networks in diffuse- and intestinal-type GC. Master regulators of the causal networks included lenvatinib, pyrotinib, histone deacetylase 1 (HDAC1), mir-196, and erb-b2 receptor tyrosine kinase 2 (ERBB2). The analysis of the HDAC1-interacting network identified the involvement of regulation of EMT by growth factors pathway, coronavirus pathogenesis pathway, and vorinostat. The network had RNA-RNA interactions with microRNAs such as mir-10, mir-15, mir-17, mir-19, mir-21, mir-217, mir-223, mir-25, mir-27, and mir-34. These molecular networks revealed in the study may lead to the identification of drug targets for GC.

This research aims to elucidate the organizational patterns of interregional economic interdependence to enhance our comprehension of the national economy's structure at a regional scale. Employing a multilayer network model, this study represents economic interdependence among Indonesian regions, utilizing the InterRegional Input-Output (IRIO) table. Through the application of various metrics, such as degree and strength distribution, assortativity coefficient, and global and local rich club coefficient, to the multilayer IRIO network, we uncover the organizational patterns of economic exchanges between provinces and economic sectors within Indonesia. Our findings demonstrate that a multilayer network approach reveals the heterogeneous and complex structure of the national economy at the regional level. By analyzing the assortativity pattern and global rich-club coefficient, we illustrate that the IRIO network exhibits a hierarchical organization, where significant provincial-sector nodes are interconnected and form dense rich clubs, extending from a few structural cores to peripheral regions. Additionally, we identify distinct connectivity patterns of non-rich nodes based on their incoming and outgoing relations. The insights gained from this study have implications for the macro-control of regional development.

Selecting road networks in cartographic generalization has consistently posed formidable challenges, driving research towards the application of intelligent models. Despite previous efforts, the accuracy and connectivity preservation in these studies, particularly when dealing with road types of similar sample sizes, still warrant improvement. To address these shortcomings, we introduce a Heterogeneous Graph Attention Network (HAN) for road selection, where feature masking method is initially utilized to assess the significance of road features. Concentrating on the most relevant features, two meta-paths are introduced within the HAN framework: one for aggregating features of the same road type within the first-order neighborhood, emphasizing local connectivity, and another extending this aggregation to the second-order neighborhood, capturing a broader spatial context. For a comprehensive evaluation, we introduce a set of metrics considering both quantitative and qualitative aspects of the road network. Compared to traditional neural networks, our HAN model demonstrates superior accuracy and F1-score, particularly for road types with comparable samples. Additionally, it enhances the overall connectivity of the selected network. In summary, our HAN-based method provides an advanced solution for road network selection, surpassing previous approaches in terms of accuracy and connectivity preservation.

This paper introduces a novel architecture—bidirectional optical neural network (BONN)—for providing backward connections alongside forward connections in artificial neural networks (ANNs). BONN incorporates laser diodes and photodiodes and exploits the properties of Köhler illumination to establish optical channels for backward directions. Thus, it has bidirectional functionality that is crucial for algorithms such as the backpropagation algorithm. BONN has a scaling limit of 96 × 96 for input and output arrays, and a throughput of 8.5 × 1015 MAC/s. While BONN's throughput may rise with additional layers for continuous input, limitations emerge in the backpropagation algorithm as its throughput doesn't scale with layer count. Successful BONN-based implementation of the backpropagation algorithm requires the development of fast spatial light modulator to accommodate frequent data flow changes. A two-mirror-like BONN and its cascaded extension are alternatives for multilayer emulation, and they help save hardware space and increase the parallel throughput for inference. An investigation into the application of the clustering technique to BONN revealed its potential to help overcome scaling limits and to provide full interconnections for backward directions between doubled input and output ports. BONN's bidirectional nature holds promise for enhancing supervised learning in ANNs and increasing hardware compactness.

Artificial Intelligence (AI) made its first appearance in 1956, and since then it has progressively introduced itself in healthcare system and patients’ information and care. AI functions can be grouped under the following headings: Machine Learning (ML), Deep Learning (DL), Artificial Neural Network (ANN), Convolutional Neural Network (CNN), Computer Vision (CV). Chronic limb-threatening ischemia (CLTI) represents the last stage of peripheral artery disease (PAD), and has increased over recent years, together with the rise in prevalence of diabetes and population ageing. Nowadays AI grants the possibility of developing new diagnostic and treatment solutions in the vascular field, given the possibility of accessing clinical, biological, and imaging data. By assessing the vascular anatomy in every patient, as well as the burden of atherosclerosis, and classifying the level and degree of disease, sizing and planning the best endovascular treatment, defining the perioperative complications risk, integrating experiences and resources between different specialties, identifying latent PAD, thus offering evidence-based solutions and guiding surgeons in the choice of the best surgical technique, AI challenges the role of the physician’s experience in PAD treatment.

Investigating the network response to node removal and the efficacy of the node removal strategies are related and fundamental questions in network science. Research studies proposed many node centralities based on the network structure ranking node to remove. The random walk (RW) on networks describes a stochastic process in which a walker travels among nodes. RW can be a model of transport, diffusion, and search on networks, and an essential tool for studying the importance of network nodes. In this manuscript, we propose four new measures of node centrality based on RW. Then, we compare the efficacy of the new RW node removal strategies to network dismantle with effective node removal strategies from the literature, such as betweenness and closeness node removal over synthetic and real-world networks. We evaluate the network dismantle along node removal using the size of the largest connected component (LCC). We find that, hence the betweenness nodes attack is the best strategy overall, the new node removal strategies based on RW show the highest efficacy in peculiar network topology. Specifically, RW strategies based on covering times emerge as the most effective strategy on a synthetic lattice network and two real-world road networks. Our results may be useful in selecting the best node attack strategies in a specific class of networks and in building more robust network structures.

Human activity recognition (HAR) using wearable sensors is gaining popularity in the healthcare domain, enabling continuous monitoring and timely interventions. The conventional HAR systems are centralised and need data sharing with a server for processing and model training. However, this data sharing leads to various challenges, such as privacy concerns, increased communication and storage costs, inefficient computational algorithms, data scarcity due to unwillingness to share, and difficulty achieving high accuracy and personalised models simultaneously. To overcome these challenges, this paper introduces a privacy and energy-aware federated learning (FL) framework for model training without data sharing. The proposed scheme incorporates spiking neural networks (SNNs), which are particularly suited for resource-constrained devices due to their event-driven information processing. The key idea is to synergise the strengths of SNNs and long short-term memory (LSTM) to create a hybrid model for robust HAR. The proposed hybrid spiking LSTM (S-LSTM) model uses the LSTM as an input layer to capture temporal dependencies efficiently, followed by spiking layers that replace the typical activation using leaky integrate-and-fire neurons. We employ the combination of surrogate gradient learning and backpropagation through time (BPTT) to update the weights of SNN layers in a supervised manner. The performance of the proposed model is evaluated using two publicly available datasets for both indoor and outdoor scenarios. The results show that the proposed S-LSTM performs much better than spike convolutional neural networks (S-CNN) and simple CNN, achieving the accuracy of 97% and 87% for indoor and outdoor datasets, respectively. Personalisation, also known as fine-tuning the global model using local data, is crucial in HAR. It allows the model to adapt to individual behaviour and preferences, making predictions more accurate. The results also show that personalisation improves the performance of participants up to 9% on average. Additionally, analysis with 50% random client participation reduces communication costs by 50% with minimal impact on accuracy.

A scalable optical convolutional neural network (SOCNN) based on free-space optics and Koehler illumination was proposed to address the limitations of the previous 4f correlator system. Unlike Abbe illumination, Koehler illumination provides more uniform illumination and reduces crosstalk. SOCNN allows for scaling up of the input array and the use of incoherent light sources. Hence, the problems associated with 4f correlator systems can be avoided. We analyzed the limitations in scaling the kernel size and parallel throughput and found that SOCNN can offer a multilayer convolutional neural network with massive optical parallelism.

The root system plays an irreplaceable role in plant growth. Its improvement can increase crop productivity. However, such system is still mysterious for us. The underlying mechanism has not been fully uncovered. The investigation on proteins related to the root system is an important means to complete this task. In the previous time, lack of root-related proteins makes it impossible to adopt machine learning methods for designing efficient models for the discovery of novel root-related proteins. Recently, a public database on root-related proteins was set up and machine learning methods can be applied in this field. In this study, we proposed a machine learning based model, named Graph-Root, for identification of root-related proteins. The features derived from protein sequences and one network were extracted, where the former features were processed by graph convolutional neural network and multi-head attention, and the later features abstracted the linkage between proteins. These features were fed into the fully connected layer to make prediction. The 5-fold cross-validation and independent tests suggested its good performance. It also outperformed the only one previous model, SVM-Root. Furthermore, the importance of each feature type and component in the proposed model was investigated.

With the rapid development of Internet of Things (IoT)-based near-earth remote sensing technology, the problem of network intrusion for near-earth remote sensing systems has become more complex and large-scale. Therefore, it is essential to seek an intelligent, automated, and robust network intrusion detection method. In recent years, network intrusion detection methods based on graph neural networks (GNNs) have been proposed. However, there are still some practical issues with these methods. For example, they have not taken into consideration the characteristics of near-earth remote sensing systems, the state of the nodes, and the temporal features. Therefore, this article analyzes the characteristics of existing near-earth remote sensing systems and proposes a spatio-temporal graph attention network (N-STGAT) that considers the state of nodes. The proposed network applies spatiotemporal graph neural networks to the network intrusion detection of near-earth remote sensing systems and validates the effectiveness of the proposed method on the latest flow-based dataset.

With the development of deep learning technology, various structures and research methods for super-resolution restoration of natural images and document images have been introduced. In particular, a number of recent studies have been conducted and developed in image restoration using generative adversarial network. Super-resolution restoration is ill-posed problem because of some complex restraints such as a lot of high-resolution images being restored for the same low-resolution image and also difficulty in restoring noises like edges, light smudging, and blurring. In this study, we utilized the spatially adaptive de-normalization (SPADE) structure for document image restoration to solve previous problems such as edge unclearness, hardness to catch features of texts, and the image color transition. Consequently, it can be confirmed that the edge of the character and the ambiguous stroke are restored more clearly when contrasting with the other previously suggested methods. Also, the proposed method’s PSNR and SSIM scores are geting 8% and 15% higher, respectively, compared to the previous methods.

Many complex networks have a connectivity that might be only partially detected or that tends to grow over time, hence the prediction of non-observed links is a fundamental problem in network science. The aim of topological link prediction is to forecast these non-observed links by only exploiting features intrinsic to the network topology. It has a wide range of real applications, like suggesting friendships in social networks or predicting interactions in biological networks.The Cannistraci-Hebb theory is a recent achievement in network science that includes a theoretical framework to understand local-based link prediction on paths of length n. In this study we introduce two innovations: theory of modelling (science) and theory of realization (engineering). For the theory of modelling we first recall a definition of network automata as a general framework for modelling the growth of connectivity in complex networks. We then show that several deterministic models previously developed fall within this framework and we introduce novel network automata following the Cannistraci-Hebb rule. For the theory of realization, we present how to build adaptive network automata for link prediction, which incorporate multiple deterministic models of self-organization and automatically choose the rule that better explains the patterns of connectivity in the network under investigation. We compare Cannistraci-Hebb adaptive (CHA) network automaton against state-of-the-art link prediction methods such as structural perturbation method (SPM), stochastic block models (SBM) and artificial intelligence algorithms for graph embedding. CHA displays an overall higher link prediction performance across different evaluation frameworks on 1386 networks. Finally, we highlight that CHA offers the key advantage to explicitly explain the mechanistic rule of self-organization which leads to the link prediction performance, whereas SPM and graph embedding not. In comparison to CHA, SBM unfortunately shows irrelevant and unsatisfactory performance demonstrating that SBM modelling is not adequate for link prediction in real networks.

Unmanned Aerial Vehicles (UAVs) in particular multirotors are becoming the {\it de facto} tool for aerial sensing and remote inspection. In large industrial facilities, a UAV can transmit an online video stream to inspect difficult to access structures, such chimneys, deposits and towers. However, the communication range is limited, constraining the UAV operation range. This limitation can be overcome with relaying UAVs placed between the source UAV and the control station, creating a line of communication links. In this work we assume the use of a digital data packet network technology, namely WiFi, and tackle the problem of defining the exact placement for the relaying UAVs that creates an end-to-end channel with maximal delivery of data packets. We consider asymmetric communication links and we show an increase as large as $15$\% in end-to-end packet delivery ratio when compared to an equidistant placement. We also discuss the deployment of such a network and propose a fully distributed method that converges to the global optimal relay positions taking, on average, 1.4 the time taken by a centralized method.

Corynebacterium glutamicum is a Gram-positive bacterium found in soil where the condition changes demand plasticity of the regulatory interactions, which study at the global scale has been challenged by the lack of data integration. Here, we update the manually-curated C. glutamicum transcriptional regulatory network, now including protein-protein interactions having a direct effect on gene transcription. The network model with regulations supported by any experimental evidence increased by 557 interactions regarding the previous (2018) version. 73 interactions supported by directed experiments were also included in a second model. We included 545 sRNA-mediated regulations in a third model with a total of 5164 interactions. We deposited the three network models in Abasy Atlas v2.4. We study the C. glutamicum regulatory structure by comparing it against the networks for more than 40 species, finding it to contrast in several global structural properties. We analyze the system-level components of the networks, finding that the inclusion of the sRNAs regulations changes their proportions, transferring part of the basal machinery to the modular class and increasing the number of modules while decreasing their size. Finally, we use strong networks of three model organisms to provide insights in future directions of the C. glutamicum network characterization.

The proliferation of mobile and IoT devices, coupled with the advances in the wireless communication capabilities of these devices, have urged the need for novel communication paradigms for such heterogeneous hybrid networks. Researchers have proposed opportunistic routing as a means to leverage the potentials offered by such heterogeneous networks. While several proposals for multiple opportunistic routing protocols exist, only a few have explored fuzzy logic to evaluate wireless links status in the network to construct stable and faster paths towards the destinations. We propose FQ-AGO, a novel Fuzzy Logic Q-learning Based Asymmetric Link Aware and Geographic Opportunistic Routing scheme that leverages the presence of long-range transmission links to assign forwarding candidates towards a given destination. The proposed routing scheme utilizes fuzzy logic to evaluate whether a wireless link is useful or not by capturing multiple network metrics, the available bandwidth, link quality, node transmission power, and distance progress. Based on the fuzzy logic evaluation, the proposed routing scheme employs a Q-learning algorithm to select the best candidate set toward the destination. We implement FQ-AGO on the NS-3 simulator and compare the performance of the proposed routing scheme with three other relevant protocols: AODV, DSDV, and GOR. For precise analysis, we consider various network metrics to compare the performance of the routing protocols. Our simulation result validates our analysis and demonstrates remarkable performance improvements in terms of total network throughput, packet delivery ration, and end-to-end delay.

Speech recognition, also known as automatic speech recognition (ASR), is a technology that enables software to transcribe spoken language into text. However, existing Amharic ASR methods require multiple separate blocks, such as language, acoustic, and pronunciation models with dictionaries, which can be time-consuming and influence performance. This study proposes an approach that replaces much of the speech pipeline with a single recurrent neural network (RNN) architecture. Our proposed architecture is based on a hybrid approach that combines a convolutional neural network (CNN) with a recurrent neural network (RNN) and a connectionist temporal classification (CTC) loss function. We conducted several experiments with noisy audio data that contain 20,000 valid sentences. The model was evaluated using the word error rate (WER) metric, achieving impressive results of 7% WER on noisy data. This approach has significant implications for the field of speech recognition, as it reduces the human effort required to create dictionaries and improves the efficiency and accuracy of ASR systems, making them more practical for real-world applications.

This paper presents a novel approach to financial network analysis by leveraging PolyModel theory. Traditional financial networks often rely on correlation matrices to represent relationships between assets, but these fail to capture the complex, non-linear interactions prevalent in financial markets. In response, we propose a method that quantifies the relationship between financial time series by comparing their reactions to a broad set of environmental risk factors. This method constructs a network based on the inherent similarities in how assets respond to external risks, offering a more robust representation of financial markets. We introduce several network topological properties, such as eigenvalues, degree, and clustering coefficients, to measure market stability and detect financial instabilities. These metrics are applied to a real-world dataset, including the DOW 30, to predict market drawdowns. Our results indicate that this PolyModel-based network framework is effective in capturing downside risks and can predict significant market drawdowns with high accuracy. Furthermore, we enhance sensitivity to smaller market changes by introducing new time-series indicators, which further improve the predictive power of the model.

In detecting DDoS, deep learning faces challenges and difficulties such as high computational demands, long training times, and complex model interpretation. This research focuses on overcoming these challenges by proposing an effective strategy for detecting DDoS attacks in unbalanced network environments. This research uses SMOTE to increase the class distribution of the data set by allowing models using LSTM to learn time anomalies effectively when DDoS attacks occur. The experiments carried out have shown significant improvement in the performance of the LSTM model when integrated with SMOTE. These include validation loss results of 0.048 for LSTM SMOTE and 0.1943 for LSTM without SMOTE, accuracy of 99.50 and 97.50. Apart from that, there was an increase in the f1 score from 93.4% to 98.3%. In this research, it is proven that SMOTE can be used as an effective strategy to improve model performance in detecting DDoS attacks on heterogeneous networks, as well as increasing model robustness and reliability.

Sudden cardiac arrest (SAC) is a leading cause of mortality. In recent years, though, many lives have been saved through the use of defibrillators by random bystanders. As a result, defibrillators have been popping up in various locations. In this work we propose a state mandate demanding the presence of a defibrillator at every pharmacy store and argue that it could greatly enhance and augment the impact of the defibrillator network. We support our claim via detailed calculations of the immediate (a defibrillator readily available) and partial (a defibrillator available at a reasonable distance) population coverage achievable by the implementation of our proposal, taking the Argolis area in Peloponnese Greece as a case study. Our findings indicate that, in addition to other benefits, the use of pharmacies to establish a defibrillator network could bring an eight-fold increase in the immediate population coverage of defibrillators and would provide immediate / partial access to a defibrillator to 18 / 24 incidents of SCA per year that would otherwise lead to death, potentially leading to the saving of numerous lives. Alternative scenaria are also discussed.

In the pursuit of energy efficiency and reduced environmental impact, adequate ventilation in enclosed spaces is essential. This study presents a hybrid neural network model designed for real-time monitoring and prediction of environmental variables. The system comprises two phases: An IoT hardware-software platform for data acquisition and decision-making, and a hybrid model combining short-term memory and convolutional recurrent structures. The results are promising and hold potential for integration into parallel processing AI architectures.

By framing the problem of determining the state of Schrodinger’s cat as a problem of searching for Hidden Markov Model (HMM), this work shows that Schrodinger’s cat is dead with stationary probabilities P(Cat = Dead) = 1 and P(Cat = Alive) = 0 consistently in various observation-based modelling of the HMM.

In the landscape of network management, software-defined networking (SDN) technology emerges as a dynamic approach, facilitating efficient network configuration for improved performance and monitoring, akin to the agility of cloud computing rather than traditional methods. However, its centralized structure exposes SDN to various attack vectors. Distributed Denial of Service (DDoS) attacks, particularly, pose a significant threat to SDN. This study employs machine learning algorithms and Network Traffic Classification Analysis (NCA) to classify SDN traffic as potential attacks. Notably, Decision Trees (DT) out shine other algorithms with a flawless 100% classification success, spotlighting their supremacy. Through this research, a robust path toward fortified SDN security takes form, where technological prowess and strategic intelligence unite for enhanced defense.