The question of molecular similarity is core in cheminformatics, and is usually assessed via a pairwise comparison based on vectors of properties or molecular fingerprints. We recently exploited variational autoencoders to embed 6M molecules in a chemical space, such that their (Euclidean) distance within the latent space so formed could be assessed within the framework of the entire molecular set. However, the standard objective function used did not seek to manipulate the latent space so as to cluster the molecules based on any perceived similarity. Using a set of some 160,000 molecules of biological relevance, we here bring together three modern elements of deep learning to create a novel and disentangled latent space, viz transformers, contrastive learning, and an embedded autoencoder. The effective dimensionality of the latent space was varied such that clear separation of individual types of molecules could be observed within individual dimensions of the latent space. The capacity of the network was such that many dimensions were not populated at all. As before, we assessed the utility of the representation by comparing clozapine with its near neighbours, and did the same for various antibiotics related to flucloxacillin. Transformers, especially when as here coupled with contrastive learning, effectively provide one-shot learning, and lead to a successful and disentangled representation of molecular latent spaces that at once uses the entire training set in their construction while allowing ‘similar’ molecules to cluster together in an effective and interpretable way.

Buildings have started to play a critical role in the stability and resilience of modern smart grids, leading to a refocusing of large scale energy management strategies from the supply side to the consumer side. When the buildings integrate local renewable energy generation in the form of renewable energy resources they become prosumers and this reflects into additional complexity into the operation of the interconnected complex energy systems. A class of methods of modelling the energy consumption patterns of the building have recently emerged as black-box input-output approaches with the ability to capture underlying consumption trends. These make use and require large quantities of quality data produces by non-deterministic processes underlying the energy consumption. We present an application of a class of neural networks, namely deep learning techniques for time series sequence modelling with the goal of accurate and reliable building energy load forecasting. The Recurrent Neural Network implementation uses Long Short-Term Memory layers in increasing density of nodes to quantify prediction accuracy. The case study is illustrated on four university buildings from temperate climates over one year of operation using a reference benchmarking dataset that allows replicable results. The obtained results are discussed in terms of accuracy metrics and computational and network architecture aspects and are considered suitable for further used in future in situ energy management at the building and neighbourhood levels.

The unpredictable increase in electrical demand affects the quality of theenergy throughout the network. A solution to the problem is the increase of distributedgeneration units which burn fossil fuels.While this is an immediate solution to theproblem the ecosystem gets affected by the emission of CO2.A promising solutionis the integration of Distributed Renewable Energy Sources (DRES) to the conventionalelectrical system, thus, introducing the concept of smart microgrids (SMG) that requirea safe, reliable and technically planned two-way communication system. This documentpresents a heuristic based on planning capable of providing a bidirectional communicationnear optimal route map, following the structure of an hybrid Fiber-Wireless (FiWi) with thepurpose of obtaining information of electrical parameters that help us to manage the useof energy by integrating conventional electrical system to SMG. A FiWi network is basedon the integration of wireless access and optical networks. This integration increases thecoverage and reliability at a lower cost. The optimization model is based on clusteringtechniques, through the construction of balanced conglomerates. The method is used forthe development of the clusters along with the Nearest-Neighbor Spanning Tree Algorithm(N-NST). Additionally, Optimal Delay Balancing (ODB) model will be used to minimizethe end to end delay of each grouping. In addition, the heuristic observes real designparameters such as: capacity and coverage. Using the Dijkstra algorithm, the routes arebuilt following the minimum shorter path. Therefore, this paper presents a heuristic able to plan the deployment of smart meters (SMs) through a tree-like hierarchical topology for theintegration of SMG at the lowest cost.

Even today, the automatic digitisation of scanned documents in general but especially the automatic optical music recognition (OMR) of historical manuscripts still remain an enormous challenge, since both handwritten musical symbols and text have to be identified. This paper focuses on the Medieval so-called square notation developed in the 11th-12th century, which is already composed of staff lines, staves, clefs, accidentals, and neumes, that are roughly spoken connected single notes. The aim is to develop an algorithm that captures both the neume and pitch, that is melody information that can be used to reconstruct the original writing. Our pipeline is similar to the standard OMR approach and comprises a novel staff line and symbol detection algorithm, based on deep Fully Convolutional Networks (FCN), which perform pixel-based predictions for either staff lines or symbols and their respective types. Then, the staff line detection combines the extracted lines to staves and yields an F1-score of over 99% for both detecting lines and complete staves. For the music symbol detection we choose a novel approach that skips the step to identify neumes and instead directly predicts note components (NCs) and their respective affiliation to a neume. Furthermore, the algorithm detects clefs and accidentals. Our algorithm recognises these symbols with an F1-score of over 96% if the type is ignored and predicts the true symbol sequence of a staff with a diplomatic symbol accuracy rate (dSAR) of about 87%. If only the NCs without their respective connection to a neume, all clefs, and accidentals are of interest the algorithm reaches an harmonic symbol accuracy rate (hSAR) of approximately 90%.

Technology advancements are evident in the healthcare field; numerous new opportunities and applications have emerged during the last years based on embedded intelligence and related to real-time diagnosis of medical issues, tele-care and telemedicine, remote monitoring of patients, computer-assisted “smart” transportation in case of emergencies, as well as new types of remotely controlled surgical operations. This paper aims to provide an overview of the application of embedded intelligence in the field of healthcare; it gathers the main critical requirements, related technologies and research advancements up to date, and it presents the most important challenges that are yet to be faced. The main focus is given in the development and optimization of Body Area Networks (BANs) based on advanced embedded sensing devices, the optimization of smart gateways in such networks, and the provision of holistic scalable and secure solutions in the healthcare domain. In addition, the paper presents two principal use cases that stem from the combination of novel information and communication technologies with classic healthcare practices, explaining the functional and non-functional requirements, as well as the mixed criticality characteristics of the associated systems.

Many real complex systems present multilayer structure where high-order metadata on one layer refers to dyadic data on a lower layer. Significant progresses to analyse high-order metadata under the assumption of community organization have been done. However, there are no planted communities in real-world networks, and the necessity of new frameworks to analyze high-order metadata regardless of community organization has been raised.Here, we propose to adopt hyperedge organization. Predicting ‘entanglements’ between a hyperedge and nodes scattered in the rest of the network might suggest structural or functional liaisons, without assumption of any community organization. We introduce a novel concept: hyperedge entanglement (HE), which associates to each hyperedge an entangled hyperedge, by means of a network operator that predicts significant ‘interactions at distance’ between network nodes and existing hyperedges. We also introduce a new challenge termed hyperedge entanglement prediction (HEP), and an algorithm to perform this task. We evaluated HEP performance on social, biological and synthetic data where, given only topology and hyperedges (such as communities or functional modules), the goal is to predict whether nodes not connected to a certain hyperedge might be candidates for a significant entanglement. Finally, as real application in diseasome systems biomedicine, we perform HEP on the human protein interactome to predict unknown gene entanglements with the COPD disease gene hyperedge. HEP predictions are validated by biological experiments, enlarging our understanding of molecular mechanisms behind COPD/aneurysm comorbidity.

This work addresses the energy-based source localization problem in wireless sensors networks. Instead of circumventing the maximum likelihood (ML) problem by applying convex relaxations and approximations (like all existing approaches do), we here tackle it directly by the use of metaheuristics. To the best of our knowledge, this is the first time that metaheuristics is applied to this type of problems. More specifically an elephant herding optimization (EHO) algorithm is applied. Through extensive simulations, the key parameters of the EHO algorithm are optimized such that they match the energy decay model between two sensor nodes. A detailed analysis of the computational complexity is presented, as well as performance comparison between the proposed algorithm and existing non-metaheuristic ones. Simulation results show that the new approach significantly outperforms the existing solutions in noisy environments, encouraging further improvement and testing of metaheuristic methods.

Cocoa is an important commodity crop not only to produce one of the most complex products such as chocolate from the sensory perspective, but one that commonly grows in developing countries close to the tropics. This paper presents novel techniques applied using cover photography and a novel computer application (VitiCanopy) to assess the canopy architecture of cocoa trees in a commercial plantation in Queensland, Australia. From the cocoa trees monitored, pod samples were collected, fermented, dried and grinded to obtain the aroma profile per tree using gas chromatography. The canopy architecture data were used as inputs in an artificial neural network (ANN) algorithm and the aroma profile considering six main aromas as targets. The ANN model rendered high accuracy (R = 0.82; MSE = 0.09) with no overfitting. The model was then applied to a satellite image from the whole cocoa field studied to produce canopy vigor and aroma profile maps up to the tree-by-tree scale. The tool developed could aid significantly the canopy management practices in cocoa trees that have a direct effect on cocoa quality.

We consider a random dynamical system arising in a model of associative memory. This system can be seen as a small (stochastic and deterministic) perturbation of a determinstic system having two weak attractors which are destroyed after the perturbation. We show, with a computer aided proof, that the system has a kind of chaotic itineracy. Typical orbits are globally chaotic, while they spend relatively long time visiting attractor's ruins.

Recent advances in recording and real-time analysis of surface electromyographic signals (sEMG) have fostered the use of sEMG human-machine interfaces for controlling personal computers, prostheses of upper limbs, and exoskeletons among others. Despite a relatively high mean performance, sEMG-interfaces still exhibit strong variance in the fidelity of gesture recognition among different users. Here we systematically study the latent factors determining the performance of sEMG-interfaces in synthetic tests and in an arcade game. We show that the degree of muscle cooperation and the amount of the body fatty tissue are the decisive factors in synthetic tests. Our data suggest that these factors can only be adjusted by a long-term training, which promotes fine-tuning of low-level neural circuits driving the muscles. A short-term training has no effect on synthetic tests, but significantly increases the game scoring. This implies that it works at a higher decision-making level, not relevant for synthetic gestures. We propose a procedure that enables quantification of the gestures’ fidelity in a dynamic gaming environment. For each individual subject the approach allows identifying “problematic” gestures that decrease gaming performance. This information can be used for optimizing the training strategy and for adapting the signal processing algorithms to individual users, which could be a way for a qualitative leap in the development of future sEMG-interfaces.

Alzheimer's is a disease for which there is no cure. Diagnosing Alzheimer's Disease (AD) early facilitates family planning and cost control. The purpose of this study is to predict the presence of AD using socio-demographic, clinical, and Magnetic Resonance Imaging (MRI) data. Early detection of AD enables family planning and may reduce costs by delaying long-term care. Accurate, non-imagery methods also reduce patient costs. The Open Access Series of Imaging Studies (OASIS-1) cross-sectional MRI data were analyzed. A gradient boosted machine (GBM) predicted the presence of AD as a function of gender, age, education, socioeconomic status (SES), and Mini-Mental State Exam (MMSE). A Residual Network with 50 layers (ResNet-50) predicted CDR presence and severity from MRI's (multi-class classification). The GBM achieved a mean 91.3% prediction accuracy (10-fold stratified cross validation) for dichotomous CDR using socio-demographic and MMSE variables. MMSE was the most important feature. ResNet-50 using image generation techniques based on an 80% training set resulted in 98.99% three class prediction accuracy on 4,139 images (20% validation set) at Epoch 133 and nearly perfect multi-class predication accuracy on the training set (99.34%). Machine Learning methods classify AD with high accuracy. GBM models may help provide initial detection based on non-imagery analysis, while ResNet-50 network models might help identify AD patients automatically prior to provider review.

The exploration and identification of targets and pathways for Atopic dermatitis (AD) treatment and diagnosis are critical for AD control. The conventional target exploration approach such as the literature review is not satisfying in terms of efficiency and accuracy. Recently, the bioinformatic approach is drawing attention for its unique advantage of high-volume data analysis for target and pathway exploration; Open Targets Platform is the targets source for this study to extract top 200 high-rank proteins from 3122 AD associated proteins. STRING, Cytoscape, CytoHubba, ClueGo, and CluePedia function had been applied for data analysis. The KEGG Mapper search & colour pathway was the pathway map resource for identified pathways; 23 key hub genes (VDR, KIT, BCL2L11, NFKBIA, KRAS, IL13, JAK2, STAT3, IL21, IL4R, REL, PDGFRB, FOXP3, RARA, RELB, EGFR, IL21R, MYC, CREBBP, NR3C1, IL2, JAK1, and KITLG). Additionally, 8 correlated pathways and the biological process had been identified; Through this study, a viable approach for target and pathway exploration had been presented. The identified AD targets and pathways will be tested for upcoming research for traditional Chinese medicinal herb interactions

The nature of contemporary Spatial Data Infrastructures lies in the provision of geospatial information in an on-demand fashion. Though recent applications identified the need to react to real-time information in a time-critical way. In particular, research efforts in the field of geospatial Internet of Things have identified substantial gaps in this context, ranging from a lack of standardization for event-based architectures to the meaningful handling of real-time information as ''events''. This manuscript presents work in the field of Event-driven Spatial Data Infrastructures with a particular focus on sensor networks and the devices capturing in-situ measurements. The current landscape of Spatial Data Infrastructures is outlined and used as the basis for identifying existing gaps that retain certain geospatial applications from using real-time information. We present a selection of approaches - developed in different research projects - to overcome these gaps. Being designed for specific application domains, these approaches share commonalities as well as orthogonal solutions and can build the foundation of an overall Event-driven Spatial Data Infrastructure.

This paper is about to estimate the suspended sediment transport amount in the streams flowing into the Sapanca Lake Basin. There are 12 subsidiary streams flowing into the Sapanca Lake Basin. With the aim of estimating the suspended sediment transport in 2012-204 in these subsidiary streams, measurements belonging the parameters such as level, cross sectioning, flow rate, temperature and suspended sediment were made monthly. Along the measurement period, weather conditions were above seasonal normal and precipitations decreased. In order to estimate the suspended sediment amount by using results of the measurement obtained, Artificial Neuron Network (ANN), Sediment Rating Curve (SRC) and Multiple Linear Regression (MLR) models were used for different scenarios. It was seen that artificial neuron networks yielded the most accurate results among the models.

In recent years, the field of deep learning achieved considerable success in pattern recognition, image segmentation and may other classification fields. There are a lot of studies and practical applications of deep learning on images, video or text classification. In this study, we suggest a method for changing the architecture of the most performing CNN models with the aim of designing new models to be used as stand-alone networks or as a component of an ensemble. We propose to replace each activation layer of a CNN (usually a ReLu layer) by a different activation function stochastically drawn from a set of activation functions: in this way the resulting CNN has a different set of activation function layers.

Plant leaf diseases can affect the plants’ leaves to an extent that the plants can collapse and die completely. These diseases may drastically drop the supply of vegetables and fruits to the market, and result in a low agricultural economy. In the literature, different laboratory methods of plant leaf disease detection have been used. These methods were time consuming and could not cover large areas for the detection of leaf diseases. This study infiltrates through the facilitated principles of the Convolutional Neural Networks (CNN) in order to model a network for image recognition and classification of these diseases. Neuroph was used to perform the training of a CNN network that recognized and classified images of the maize leaf diseases that were collected by use of a smart phone camera. A novel way of training and the methodology used, expedite a quick and easy implementation of the system in practice. The developed model was able to recognize 3 different types of maize leaf diseases out of healthy leaves. The Northern Corn Leaf Blight (Exserohilum), Common Rust (Puccinia sorghi) and Gray Leaf Spot (Cerospora) diseases were chosen for this study as they affect most parts of Southern Africa’s maize fields.

Communication between devices has transitioned from wired to unwired. Wireless networks have been in use widely around the globe since the advent of smartphones, IoT devices and other technologies that are compatible with wireless mode of communication. At the same time security issues have also increased in such communication methods. The aim of this paper is to propose security and privacy issues of the wireless networks and present them through comprehensive surveys. In context of security issues, there are 2 typical DDoS attacks - HTTP flood and SYN flood. Other than DDoS attacks, there are several other threats to wireless networks. One of the most prevalent include security issues in Internet of Things. In terms of privacy issues in a wireless network, location-based applications, individual data, cellular network and V2G (Vehicle to Grid) network are surveyed. The survey is hosted using questionnaire and responses of 70 participants is recorded. It is observed from the survey results that many groups of people lack the knowledge of security and privacy of wireless technologies and networks despite their increased use, however, students are relatively more aware and have strong knowledge of those issues. It is concluded from the results that an effective solution to these problems can be hosting campaigns for spreading the security and privacy laws to help the groups of people who are lagging behind in this domain of knowledge become more aware. A unique solution is also presented to overcome the security issues which include implementation of detection and mitigation techniques, implementing Blockchain in the IoT devices and implementing fog computing solutions. The unique solutions to overcome the privacy issues are proposed in the form of a privacy approach from the LBS server between pairs of users to increase the implementation of DSPM and blockchain as a solution.

Over the past decade, recognition of human activities (HAR) has become a vibrant field of research, in particular, the spread in our everyday lives of electronics such as mobile phones, smart cell phones, and video cameras. Furthermore, the advancement in the field of deep methodologies and other paradigms have enabled scientists to enable HAR in many areas, consisting of activities in fitness and wellness. For instance, HAR is one of many resorts to support older people through day-to-day activities to support their cognition and physicality. This study is centered on the key aspects deep learning plays in the development of HAR applications. Although numerous HAR examination studies were carried out previously, there have been no overall studies on this subject, in all the earlier studies there were only specific HAR-related subjects. A detailed review covering all the main subjects in this area is therefore essential. This study discusses the latest developments and works in HAR. It separates the methods and the advantages and disadvantages of each method group. This paper finally addresses many problems in the current HAR subject and provides recommendations for potential study.

In this paper the authors describe the architecture of a multidisciplinary data acquisition and visualization platform devoted to the management of coastal environments. The platform integrates heterogeneous data acquisition sub-systems that can be roughly divided in two main categories: remote sensing systems and in-situ sensing systems. Remote sensing solutions include aerial and underwater remote data acquisition while in-situ sensing solutions include the use of RFID tracers, Wireless Sensor Networks and imaging techniques. All the data collected by these subsystems are stored, integrated and fused on a single platform that is also in charge of data visualization. This last task is carried out according to the paradigm of Augmented Virtuality which foresees the augmentation of a virtually reconstructed environment with data collected in the real world. The described solution proposes a novel holistic approach where different disciplines concur, with different data acquisition techniques, to a large scale definition of coastal dynamics, in order to better describe and face the coastal erosion phenomenon. The overall framework has been conceived by the so-called Team COSTE, a joint research team between the Universities of Pisa, Siena and Florence.

Unmanned Aerial Vehicles (UAVs) severing as the relay is an effective technology method to extend the coverage. It can also alleviate the congestion and increase the throughput, especially applied in UAV networks. However, since the energy of UAVs is limited and the resources in UAV networks are scarce, how to optimize the network delay performance under these constraints should be well investigated. Besides, the relationship among different resources, e.g. power and bandwidth, is coupled which makes the optimization more complex. This article investigates the problem of joint power and bandwidth allocation in UAV backhaul networks, which considers both the delay performance and the resource utilization efficiency. Considering the heterogeneous locations characteristics of different UAVs, we formulate the optimization problem as a Stackelberg game. The relay UAV acts as the leader and extended UAVs act as followers. Their utility functions take both the delay durance and the resource consumption into account. To capture the competitive relationship among followers, the sub-game is proved to be an exact potential game and exists Nash equilibriums (NE). The Stackelberg Equilibrium (SE) is proved afterwards. We utilize a hierarchical learning algorithm (HLA) to find out the best resource allocation strategies, which also reduces the computational complexity. Simulation results demonstrate the effectiveness of the proposed method.

A numerical approach for generating a limited number of water demand scenarios and estimating their occurrence probabilities in a Water Distribution Network (WDN) is proposed. This approach makes use of the demand scaling laws in order to consider the natural variability and spatial correlation of nodal consumptions. The scaling laws are employed to determine the statistics of nodal consumption as a function of the number of users and the main statistical features of the unitary user's demand. Besides, consumption at each node is considered to follow a Gamma probability distribution. A high number of groups of cross-correlated demands, i.e., scenarios, for the entire network were generated using Latin Hypercube Sampling (LHS) and the numerical procedure proposed by Iman and Conover. The Kantorovich distance is used to reduce the number of scenarios and estimate their corresponding probabilities, while keeping the statistical information on nodal consumptions. By hydraulic simulation, the whole number of generated demand scenarios was used to obtain a corresponding number of pressure scenarios on which the same reduction procedure was applied. The probabilities of the reduced scenarios of pressure were compared with the corresponding probabilities of demand.

Scoring targets in shooting sports is a crucial and time-consuming task that relies on manually counting bullet holes. This paper introduced an automatic score detection model using object detection techniques. The performance of seven models belonging to two different architectural setups was compared. Models like YOLOv8n, YOLOv8s, YOLOv8m, RetinaNet-50, and RetinaNet-101 are single-shot detectors, while Faster RCNN-50 and Faster RCNN-101 belong to the two-shot detectors category. The dataset was manually captured from the shooting range and expanded by generating more versatile data using Python code. Before the dataset was trained to develop models, it was resized (640x640) and augmented using Roboflow API. The trained models were then assessed on the test dataset, and their performance was compared using matrices like mAP50, mAP50-90, precision, and recall. The results showed that YOLOv8 models can detect multiple objects with good confidence scores. Among all, YOLOv8m performed the best with the highest mAP50 value of 96.7%, followed by the performance of YOLOv8s with the mAP50 value of 96.5%. It is suggested that if the system is to be implemented in a real-time environment, YOLOv8s is a better choice since it took significantly less inference time (2.3ms) than YOLOv8m (5.7ms) and yet generated the competitive mAP50.

Systems biology has been established for more than a decade in the post-genomic era. With the help of the computational and mathematical tools, systems biology reconstitutes the entire scenario of the cell, tissue and even organism from the pieces data generated in the past decades. However, the modern biology is mainly focusing on the structure and function of the biomolecule, cell, tissue or organ, which are far from the essence of the life because of missing thermodynamic information. It is doubtable that the current systems biology-based omics is no-how to fully understand the dynamic courses of the structure, function and information in life. For this reason, we promote a novel concept of aquamoleculomics, in which the biological structure and function as well as thermodynamic characteristics and bioinformation of the aquamolecule complexes are included in this theoretical model of systems biology. Water is mother of life, matter and matrix of organism. Indeed, the fundamental roles of H₂O molecules in biological processes might be dramatically underestimated. Extremely speaking, H₂O networks in the living system might be engaged in all the biological processes including building all the biological structures, the residential places of the motherhood molecules as the honeycombs of honeybees.

Artificial Intelligence (AI) refers to the simulation of human intelligence in machines, using machine learning, deep learning and neural networks. AI enables machines to learn from experience and perform human-like tasks. The field of AI research has been developing fast over the past five to ten years, due to the rise of ‘big data’ and increasing computing power. In the medical area, AI can be used to improve diagnosis, prognosis, treatment, surgery, drug discovery, or for other applications. Therefore, both academia and industry are investing a lot in AI. This review investigates the biomedical literature (in the PubMed and Embase databases) by looking at bibliographical data, observing trends over time and occurrences of keywords. Some observations are made: AI has been growing exponentially over the past few years; it is used mostly for diagnosis; COVID-19 is already in the top-5 of diseases studied using AI; the United States, China, United Kingdom, South Korea and Canada are publishing the most articles in AI research; MIT is the world’s leading university in AI research; and convolutional neural networks are by far the most popular deep learning algorithms at this moment. These trends could be studied in more detail, by studying more literature databases or by including patent databases. More advanced analyses could be used to predict in which direction AI will develop over the coming years. The expectation is that AI will keep on growing, in spite of stricter privacy laws, more need for standardization, bias in the data, and the need for building trust.

DARPA, the Defense Advanced Research Projects Agency from the United States, has started the Spectrum Collaboration Challenge with the aim to encourage research and development of coexistence and collaboration techniques of heterogeneous networks in the same wireless spectrum bands. Team SCATTER has been participating in the challenge since its beginning, back in 2016. SCATTER’s open-source software-defined physical layer (SCATTER PHY) has been developed as a standalone application, with the ability to communicate with higher layers of SCATTER’s system via ZeroMQ, and uses USRP X310 software-defined radio devices to send and receive wireless signals. SCATTER PHY relies on USRP’s ability to schedule timed commands, uses both physical interfaces of the radio devices, utilizes the radio’s internal FPGA board to implement custom high-performance filtering blocks in order to increase its spectral efficiency as well as enable reliable usage of neighboring spectrum bands. This paper describes the design and main features of SCATTER PHY and showcases the experiments performed to verify the achieved benefits.

In the fifth generation communication system, millimeter wave (mmWave) networks will coexist with traditional micro wave (μWave) networks, which allows for higher data transmission rate and better user experience. In this paper, we give a comprehensive framework of mathematical models and analytical methods for multi-tier mmWave and μWave hybrid caching networks based on stochastic geometry. We propose an association strategy by assuming average biased-received power association and Rayleigh fading. Accordingly, by using a D-ball approximating blockage model of mmWave networks, expressions of the cell association probability and the coverage probability of the hybrid networks are derived. Also, we use the average successful delivery probability as the performance metric to analyze the existing caching placement strategies. Simulation results validate the accuracy of our analytical conclusions.

Accurate and fast traffic flow forecasting is vital in intelligent transportation system because many of the advanced features in intelligent transportation systems are based on it. However, existing methods have poor performance regarding accuracy and computational efficiency in long-term traffic flow forecasting under big data. Hence, we propose an improved Long short-term memory (LSTM) Network and its cluster computing implementation in this paper to address the above challenge. We propose a singular point probability LSTM (SDLSTM) algorithm. The method discards the units of the network according to the singular point probability during the training process and amends the SDLSTM by Autoregressive Integrated Moving Average Model (ARIMA) to achieve the accurate prediction of 24-hour traffic flow data. Furthermore, the paper designs a scheme for implementing this method through cluster computing to shorten the calculation time and improve the system's operating speed. Theoretical analysis and experimental results show that SDLSTM gains a higher accuracy rate and better stability in the long-term traffic flow forecasting compared with previous methods.

Opportunistic routing uses the broadcasting nature of the wireless medium to increase the number of potential forwarding nodes in the network. This, in turn improves the delivery rate and reliability of data transmission in the network. Compared to all previous classes of protocols, opportunistic routing offers numerous advantages which is exploited by the latest applications for efficient communication and resource sharing in dynamic ad hoc networks. The objective of this research work is to review and classify all the major opportunistic routing protocols proposed for dynamic ad hoc networks. Further the issues and challenges with each of these existing protocol is discussed and future research directions are put forward.

The dynamics of many complex systems can be classified as ordered, chaotic, or critical. Order offers stability and robustness, while chaos allows for change and adaptability. Criticality, then, is often seen as a balance required by living systems at different scales. In classical models, however, criticality is only found near phase transitions, restricting the parameter space (and thus the likelihood) of critical dynamics, as most parameters yield ``undesirable'' solutions. Here we show that this limitation is due to the homogeneity built-in these models, i.e., all elements sharing parameter values. By exploring heterogeneous versions of archetypal models in physics and computer science, we observe critical dynamics in a broader range of parameters, and thus could be more common than previously thought.

RAS oncogenes are amongst the most commonly mutated proteins in human cancers. They regulate a wide range of effector pathways that control cell proliferation, survival, differentiation, migration and metabolic status. Including aberrations in these pathways, RAS dependent signaling is altered in more than half of human cancers. Targeting mutant RAS proteins and their downstream oncogenic signaling pathways has been elusive. However, recent results comprising detailed molecular studies, large scale omics studies and computational modeling have painted a new and more comprehensive portrait of RAS signaling that helps us to understand the intricacies of RAS, how its physiological and pathophysiological functions are regulated, and how we can target them. Here, we review these efforts particularly trying to relate the detailed mechanistic studies with global functional studies. We highlight the importance of computational modeling and data integration to derive an actionable understanding of RAS signaling that will allow us to design new mechanism based therapies for RAS mutated cancers.

This last decade, the amount of data exchanged in the Internet increased by over a staggering factor of 100, and is expected to exceed well over the 500 exabytes by 2020. This phenomenon is mainly due to the evolution of high speed broadband Internet and, more specifically, the popularization and wide spread use of smartphones and associated accessible data plans. Although 4G with its long-term evolution (LTE) technology is seen as a mature technology, there is continual improvement to its radio technology and architecture such as in the scope of the LTE Advanced standard, a major enhancement of LTE. But for the long run, the next generation of telecommunication (5G) is considered and is gaining considerable momentum from both industry and researchers. In addition, with the deployment of the Internet of Things (IoT) applications, smart cities, vehicular networks, e-health systems, and Industry 4.0, a new plethora of 5G services has emerged with very diverging and technologically challenging design requirements. These include: high mobile data volume per area, high number of devices connected per area, high data rates, longer battery life for low-power devices, and reduced end-to-end latency. Several technologies are being developed to meet these new requirements. Among these we list ultra-densification, millimeter Wave usage, antennas with massive multiple-input multiple-output (MIMO), antenna beamforming to increase spacial diversity, edge/fog computing, among others. Each of these technologies brings its own design issues and challenges. For instance, ultra-densification and MIMO will increase the complexity to estimate channel condition and traditional channel state information (CSI) estimation techniques are no longer suitable due to the complexity of the new scenarios. As a result, new approaches to evaluate network condition such as by continuously collecting and monitoring key performance indicators become necessary. Timely decisions are needed to ensure the correct operation of such network. In this context, deep learning (DL) models could be seen as one of the main tools that can be used to process monitoring data and automate decisions. As these models are able to extract relevant features from raw data (images, texts, and other types of unstructured data), the integration between 5G and DL looks promising and one that requires exploring. As main contributions, this paper presents a systematic review about how DL is being applied to solve some 5G issues. We examine data from the last decade and the works that addressed diverse 5G problems, such as physical medium state estimation, network traffic prediction, user device location prediction, self network management, among others. We also discuss the main research challenges when using DL models in 5G scenarios and identify several issues that deserve further consideration.

Exponential increase in mobile video delivery will continue with the demand for higher resolution, multi-view and large-scale multicast video services. Novel fifth generation (5G) 3GPP New Radio (NR) standard will bring a number of new opportunities for optimizing video delivery across both 5G core and radio access network. One of the promising approaches for video quality adaptation, throughput enhancement and erasure protection is the use of packet-level random linear network coding (RLNC). In this work, we discuss the integration of RLNC into the 5G NR standard, building upon the ideas and opportunities identified in 4G LTE. We explicitly identify and discuss in detail novel 5G NR features that provide support for RLNC-based video delivery in 5G, thus pointing out to the promising avenues for future research.

Studies performed at single-cell resolution have demonstrated the physiological significance of cell-to-cell variability. Various types of mathematical models and systems analyses of biological networks have further been used to gain a better understanding of the sources and regulatory mechanisms of such variability. In this work, we present a novel sensitivity analysis method, called molecular density function perturbation (MDFP), for the dynamical analysis of cellular heterogeneity. The proposed analysis is based on introducing perturbations to the density or distribution function of the cellular state variables at specific time points, and quantifying how such perturbations affect the state distribution at later time points. We applied the MDFP analysis to a model of signal transduction pathway involved in TRAIL (tumor necrosis factor-related apoptosis-inducing ligand)-induced apoptosis in HeLa cells. The MDFP analysis showed that caspase-8 activation regulates the timing of the switch-like increase of cPARP (cleaved poly(ADP-ribose) polymerase), an indicator of apoptosis. Meanwhile, the cell-to-cell variability in the commitment to apoptosis depended on mitochondrial outer membrane permeabilization (MOMP) and events following MOMP, including the release of Smac (second mitochondria-derived activator of caspases) and cytochrome-C from mitochondria, the inhibition of XIAP (X-linked inhibitor of apoptosis) by Smac and the formation of apoptosome.

This paper investigates the effects of air void topology on hydraulic conductivity in asphalt mixtures with porosity in the range 14%-31%. Virtual asphalt pore networks were generated using the Intersected Stacked Air voids (ISA) method, with its parameters being automatically adjusted by the means of a differential evolution optimisation algorithm, and then 3D printed using transparent resin. Permeability tests were conducted on the resin samples to understand the effects of pore topology on hydraulic conductivity. Moreover, the pore networks generated virtually were compared to real asphalt pore networks captured via X-ray Computed Tomography (CT) scans. The optimised ISA method was able to generate realistic 3D pore networks corresponding to those seen in asphalt mixtures in term of visual, topological, statistical and air void shape properties. It was found that, in the range of porous asphalt materials investigated in this research, the high dispersion in hydraulic conductivity at constant air void content is a function of the average air void diameter. Finally, the relationship between average void diameter and the maximum aggregate size and gradation in porous asphalt materials was investigated.

Optical networks are prone to power jamming attacks intending service disruption. This paper presents a Machine Learning (ML) framework for detection and prevention of jamming attacks in optical networks. We evaluate various ML classifiers for detecting out-of-band jamming attacks with varying intensities. Numerical results show that artificial neural network is the fastest ($10^6$ detection per second) for inference and most accurate ($\approx 100 \%$) in detecting power jamming attacks as well as identifying the optical channels attacked. We also discuss and study a novel prevention mechanism when the system is under active jamming attacks. For this scenario, we propose a novel resource reallocation scheme that utilizes the statistical information of attack detection accuracy to lower the probability of successful jamming of lightpaths while minimizing lightpaths' reallocations. Simulation results show that the likelihood of jamming a lightpath reduces with increasing detection accuracy, and localization reduces the number of reallocations required.

Predicting how a genetic change affects a given character is a major challenge in biology, and being able to tackle this problem relies on our ability to develop realistic models of gene networks. However, such models are rarely tractable mathematically. In this paper, we propose a mathematical analysis of the sigmoid variant of the Wagner gene-network model. By considering the simplest case, that is, one unique self-regulating gene, we show that numerical simulations are not the only tool available to study such models: theoretical studies can be done too, by mathematical analysis of discrete dynamical systems. It is first shown that the particular sigmoid function can be theoretically investigated. Secondly, we provide an illustration on how to apply such investigations in the case of the dynamical system representing the one self-regulating gene.

This paper aims to analyze the effects of Covid-19 on financial markets from the perspective of Complexity. The Covid-19 pandemic has caused turmoil in financial markets and is already one of the most important financial crises in history, causing a fall in several stock markets as well as great volatility. Unlike past crises, which in most cases were caused either by problems of fiscal deficits or in the financial system, this crisis has its origin in an epidemic disease, which occurred in Wuhan, China, and quickly spread across the globe affecting transport networks, commerce and finance, and will affect the public debt of many countries. This is a systemic situation in which there is a need to consider several interconnected systems, financial instability and high financial risk, something that econophysics and some complexity theorists already do. Therefore, this paper intends to show, theoretically, the systemic and complex character of the recent financial crisis caused by Covid-19.

Global environmental and societal changes threaten the cultures of Indigenous Peoples and Local Communities (IPLC). Despite the importance of IPLC worldviews and knowledge systems to human well-being and biodiversity, risks to these cultural resources are commonly simplified or neglected in environmental impact assessments, in part because cultural impacts are often indirect and therefore difficult to demonstrate. Here, we show that dependency of a culture on the environment can be mapped through human connections with biophysical elements in their environment. We illustrate a rich variety of cultural values that connect an indigenous Māori tribe in New Zealand with their local environment, then evaluate the resilience of this socio-environmental value system to environmental changes. Our results detail how loss of access to key environmental elements can have extensive direct and cascading impacts on multiple facets of indigenous cultural heritage. Consequently, considering only direct effects of environmental change on cultural heritage, or treating the richness of IPLC environmental relations simplistically, can severely underestimate the seriousness of environmental impacts on IPLC culture. Thus, protecting Earth’s cultural and biological diversity requires inclusion of human-environment relationships in environmental impact assessments.

Recent progress in deep learning has led to accurate and efficient generic object detection networks. Training of highly reliable models depends on large datasets with highly textured and rich images. However, in real-world scenarios, the performance of the generic object detection system decreases when (i) occlusions hide the objects, (ii) objects are present in low-light images, or (iii) they are merged with background information. In this paper, we refer to all these situations as challenging environments. With the recent rapid development in generic object detection algorithms, notable progress has been observed in the field of object detection in challenging environments. However, there is no consolidated reference to cover state-of-the-art in this domain. To the best of our knowledge, this paper presents the first comprehensive overview, covering recent approaches that have tackled the problem of object detection in challenging environments. Furthermore, we present the quantitative and qualitative performance analysis of these approaches and discuss the currently available challenging datasets. Moreover, this paper investigates the performance of current state-of-the-art generic object detection algorithms by benchmarking results on the three well-known challenging datasets. Finally, we highlight several current shortcomings and outline future directions.

The sensor nodes have limited computation, sensing, communication capabilities and generally operated by batteries in a harsh atmosphere with non-replenish able power sources. These limitations force the sensor network subject to failure because most of the energy is spent on sensing, computing and data transmission. This paper introduces an Energy Efficient Clustering and Shortest-Path Routing Protocol (EECSRP) to assist Wireless Sensor Networks (WSNs) by (a) extending the lifespan of the network (b) effectively using the battery power (c) decreasing the network overhead and (d) ensuring a high packet transmission ratio with minimal delay. The delay time-based Cluster Head (CH) is elected based on the node degree, residual energy and Received Signal Strength (RSS) to accomplish the goal. Additionally, the RSS-based network partitioning is implemented to evaluate the gradient based on demand routing between source (sensing node) and destination (BS). Whenever the current CH residual energy goes under the threshold level, the proposed protocol performs the clustering process, reducing the exchange of control packets. However, the BS periodically gathers the data from every single CH which helps to reduce the collision and Medium Access Control (MAC) layer conflict. From the simulation results, it is the evident that the proposed protocol performance in terms of average end-to-end latency, packet delivery ratio, average energy consumption and control overhead is better than the well-known current protocols.

Deep learning models are part of the family of artificial neural networks and, as such, it suffers of catastrophic interference when they learn sequentially. In addition, most of these models have a rigid architecture which prevents the incremental learning of new classes. To overcome these drawbacks, in this article we propose the Self-Improving Generative Artificial Neural Network (SIGANN), a type of end-to-end Deep Neural Network system which is able to ease the catastrophic forgetting problem when leaning new classes. In this method, we introduce a novelty detection model to automatically detect samples of new classes, moreover an adversarial auto-encoder is used to produce samples of previous classes. This system consists of three main modules: a classifier module implemented using a Deep Convolutional Neural Network, a generator module based on an adversarial autoencoder; and a novelty detection module, implemented using an OpenMax activation function. Using the EMNIST data set, the model was trained incrementally, starting with a small set of classes. The results of the simulation show that SIGANN is able to retain previous knowledge with a gradual forgetfulness for each learning sequence. Moreover, SIGANN can detect new classes that are hidden in the data and, therefore, proceed with incremental class learning.

Created in 1905, the Société des Artistes Meridionaux organized annual exhibitions in Toulouse, in order to promote regional arts and (re)created a "Latin" - and modern - style. Based on a corpus of 11,486 artworks exhibited at the SAM between 1907 and 1939, this paper seeks to measure the contours and particularities of the Salons des Artistes Meridionaux, using a computational approach.

Retinopathy is a human eye disease that causes changes in retinal blood vessels that leads to bleed, leak fluid and vision impairment. Symptoms of retinopathy are blurred vision, changes in color perception, red spots, and eye pain. In this paper, a new methodology based on Convolutional Neural Networks (CNN) is developed and proposed to diagnose and give a decision about the presence of retinopathy. The CNN model is trained by different images of eyes that have retinopathy and those which do not have retinopathy. The performance of the proposed model is compared with the related methods of DREAM, KNN, GD-CNN and SVM. Experimental results show that the proposed CNN performs better.

Currently, the use of numerical models for reproducing the landscape evolution of a river basin is part of the day-by-day research activities of fluvial engineers and geomorphologists. However, despite landscape modelling is based on a rather long tradition, and scientists and practitioners are trying to schematize the processes involved in the evolution of a landscape since decades, there is still the need for improving both the knowledge of the physical mechanisms and their numerical coding. The present review focuses on the first aspect, discussing the main components of a landscape evolution model and their more common schematizations, presenting possible open questions to be addressed towards an improvement of the reliability of such kind of models in describing the fluvial geomorphology.

Unmanned Aerial Vehicles (UAVs) are being increasingly viewed as valuable tools to aid the management of farms. This kind of technology can be particularly useful in the context of extensive cattle farming, as production areas tend to be expansive and animals tend to be more loosely monitored. With the advent of deep learning, and Convolutional Neural Networks (CNNs) in particular, extracting relevant information from aerial images has become more effective. Despite the technological advancements in drone, imaging and machine learning technologies, the application of UAVs for cattle monitoring is far from being thoroughly studied, with many research gaps still remaining. In this context, the objectives of this study were threefold: 1) to determine the highest possible accuracy that could be achieved in the detection of animals of the Canchim breed, which is visually similar to the Nelore breed (\textit{Bos taurus indicus}); 2) to determine the ideal Ground Sample Distance (GSD) for animal detection; 3) to determine the most accurate CNN architecture for this specific problem. The experiments involved 1,853 images containing 8,629 samples of animals, and 15 different CNN architectures were tested. A total of 900 models were trained (15 CNN architectures * 3 spacial resolutions * 2 datasets * 10-fold cross validation), allowing for a deep analysis of the several aspects that impact the detection of cattle using aerial images captured using UAVs. Results revealed that many CNN architectures are robust enough to reliably detect animals in aerial images even under far from ideal conditions, indicating the viability of using UAVs for cattle monitoring.

In the remote sensing domain, it is crucial to complete semantic segmentation on the raster images, e.g., river, building, forest, etc, on raster images. A deep convolutional encoder--decoder (DCED) network is the state-of-the-art semantic segmentation method for remotely sensed images. However, the accuracy is still limited, since the network is not designed for remotely sensed images and the training data in this domain is deficient. In this paper, we aim to propose a novel CNN for semantic segmentation particularly for remote sensing corpora with three main contributions. First, we propose applying a recent CNN called a global convolutional network (GCN), since it can capture different resolutions by extracting multi-scale features from different stages of the network. Additionally, we further enhance the network by improving its backbone using larger numbers of layers, which is suitable for medium resolution remotely sensed images. Second, "channel attention'' is presented in our network in order to select the most discriminative filters (features). Third, "domain-specific transfer learning'' is introduced to alleviate the scarcity issue by utilizing other remotely sensed corpora with different resolutions as pre-trained data. The experiment was then conducted on two given datasets: (i) medium resolution data collected from Landsat-8 satellite and (ii) very high resolution data called the ISPRS Vaihingen Challenge Dataset. The results show that our networks outperformed DCED in terms of $F1$ for 17.48% and 2.49% on medium and very high resolution corpora, respectively.

Cooperative non-orthogonal multi access communication is a promising paradigm for the future wireless networks because of its advantages in terms of energy efficiency, wider coverage and interference mitigating. In this paper, we study the secrecy performance of a downlink cooperative non-orthogonal multi access (NOMA) communication system under the presence of an eavesdropper node. Smart node selection based on feed forward neural networks (FFNN) is proposed in order to improve the physical layer security (PLS) of a cooperative NOMA network. The selected cooperative relay node is employed to enhance the channel capacity of the legal users, where the selected cooperative jammer is employed to degrade the capacity of the wiretapped channel. Simulations of the secrecy performance metric namely the secrecy capacity ($C_S$) are presented and compared with the conventional technique based on fuzzy logic node selection technique. Based on our simulations and discussions the proposed technique outperforms the existing technique in term the of secrecy performance.

The detection of seismic events at regional and teleseismic distances is critical to Nuclear Treaty Monitoring. Traditionally, detecting regional and teleseismic events has required the use of an expensive multi-instrument seismic array; however in this work, we present DeepPick, a novel seismic detection algorithm capable of array-like performance from a single trace. We achieve this directly, by training our single-trace detector against labeled events from an array catalog, and by utilizing a deep temporal convolutional neural network. The training data consists of all arrivals in the International Seismological Centre Catalog for seven seismic arrays over a five year window from 1 Jan 2010 to 1 Jan 2015, yielding a total training set of 608,362 detections. The test set consists of the same seven arrays over a one year window from 1 Jan 2015 to 1 Jan 2016. We report our results by training the algorithm on six of the arrays and testing it on the seventh, so as to demonstrate the transportability and generalization of the technique to new stations. Detection performance against this test set is outstanding. Fixing a type-I error rate of 1%, the algorithm achieves an overall recall rate of 73% on the 141,095 array beam picks in the test set, yielding 102,394 correct detections. This is more than 4 times the 23,259 detections found in the analyst-reviewed single-trace catalogs over the same period, and represents an 8dB improvement in detector sensitivity over current methods. These results demonstrate the potential of our algorithm to significantly enhance the effectiveness of the global treaty monitoring network.

This work has its origin in intuitive physical and statistical considerations. The problem of optimizing an artificial neural network is treated as a physical system, composed of a conservative vector force field. The derived scalar potential is a measure of the potential energy of the network, a function of the distance between predictions and targets. Starting from some analogies with wave mechanics, the description of the sys-tem is justified with an eigenvalue equation that is a variant of the Schr˜odinger equation, in which the potential is defined by the mutual information between inputs and targets. The weights and parameters of the network, as well as those of the state function, are varied so as to minimize energy, using an equivalent of the variational theorem of wave mechanics. The minimum energy thus obtained implies the principle of minimum mutual information (MinMI). We also propose a definition of the potential work produced by the force field to bring a network from an arbitrary probability distribution to the potential-constrained system, which allows to establish a measure of the complexity of the system. At the end of the discussion we expose a recursive procedure that allows to refine the state function and bypass some initial assumptions, as well as a discussion of some topics in quantum mechanics applied to the formalism, such as the uncertainty principle and the temporal evolution of the system. Results demonstrate how the minimization of energy effectively leads to a decrease in the average error between network predictions and targets.

The selection of the best mine production technical indicators is crucial to increasing a mine’s economic benefit and saving resources for sustainability. Therefore, this research proposes a ‘multi-objective optimization model’ based on a ‘fast and elitist Non-dominated Sorting Genetic Algorithm’ (NSGA-II) and ‘Artificial Neural Networks’ (ANN) for the optimization of production technical indicators in the entire geology, mining and beneficiation metal mine production processes. The multi-objective optimization model has decision variables including ‘cut-off grade,’ ‘industrial grade’ and ‘loss rate,’ with objectives being ‘economic benefit (profit)’ and ‘resource benefit (metal volume).’ First, the relationship between the technical indicators of mine production is studied. The REG model, MATLAB’s own ksdensity function and the BP neural network are used to calculate the ore weight, the probability density of grade distribution, the dilution rate, the concentration ratio and the concentrate grade, and to further calculate geological reserves, profit and metal volume. Then, the NSGA-II is applied to maximize profit and metal volume simultaneously. Finally, the model is applied to the Huogeqi copper mine. The optimization result is a set of multiple optimal solutions called Pareto optimal solutions. Compared with the plan data, the profit and metal volume of partial optimization results increased by 2.89% and 2.64% simultaneously. These Pareto optimal solutions can help decision makers in bettering the actual process of metal mine production.

Opportunistic routing has emerged as one of the prominent solutions for efficient data delivery in ad hoc networks especially in highly dynamic scenarios. Opportunistic routing uses the broadcasting nature of the wireless medium to increase the number of potential forwarding nodes in the network, thereby improving the delivery rate. But most of the proposed opportunistic routing protocols suffer from increased duplicate data forwarding at the intermediate nodes in the network which leads to increase in traffic and reduced efficiency. We propose a novel method which uses acknowledgements for controlling the data forwarding by intermediate nodes and thereby eliminating most of the duplicate forwarding in the network. Simulation results show that our proposed method is much more efficient that all the major existing opportunistic protocols.

The need for the most accurate and flexible system of revenue collection from internal sources has become a matter of extreme urgency and importance in e-governance. This need underscores the eagerness on the part of the Government to look for a new principle and policy of revenue collection or to become aggressive and innovative in the mode of collecting revenue from existing sources using the present system. The Boards of some Governments in Africa, even up to the moment are facing a lot of setbacks in performing their tasks due to the manual system of revenue collection from the public. This can be improved through an effective collection of revenue using the most accurate and flexible system. Tax is usually collected in the form of specific sales tax, general sales tax, corporate income tax, individual income tax, property tax and inheritance tax. Problems such as high cost of collection, fraud, underpayment, leakage in revenue, poor access to information, poor tracking of defaulters is at the increase. As a result of this, there is need to computerize the revenue collection system. Computerized systems have proven to introduce massive efficiencies and quick collection of revenue from the public. This research work demonstrates how to design and implement an automated system of revenue collection and how to maintain a secured database for collected tax information. This research delves into the study of how machine learning algorithms and Software-defined Networks improve the security of such automated systems.

This work describes and compares the backpropagation algorithm with the Extended Kalman filter, a second-order training method which can be applied to the problem of learning neural network parameters and is known to converge in only a few iterations. The algorithms are compared with respect to their effectiveness and speed of convergence using simulated data for both, a regression and a classification task.

The introduction of social commerce ushered in a new era in business-consumer interaction. As a result, more power has passed from the vendor to the buyer, primarily fueling e-commerce acceptance. As a result, understanding consumer behaviour in the context of social commerce adoption has become essential for businesses looking to persuade customers by using the power of social ties and support.Furthermore, such social ties will facilitate trust as the most promising benefit while reducing perceived risk, which has always been a critical problem with online commerce. This study proposes a paradigm for understanding the impact of social commerce on the stages of the consumer decision-making process: need recognition, information search, alternative evaluation, purchase decision, and post-purchase behaviour, with a focus on social support. In this respect, relevant literature in the subject of social commerce either (1) lacks an adequate explanatory model, (2) has a solid theoretical base, or (3) contains practically complex theories with insufficient empirical data. The research model applies the Social Commerce Constructs (SCC): recommendations and referrals, forums and communities, and ratings and reviews to study the respective influence on the consumer decision-making process phases. This paper aims to understand the influence of social commerce on an integrative model that incorporates all customer choice phases while expecting new knowledge. Furthermore, it is advised that this conceptual model be empirically verified to evaluate the practical consequences.

In this paper, we address the problem of car detection from aerial images using Convolutional Neural Networks (CNN). This problem presents additional challenges as compared to car (or any object) detection from ground images because features of vehicles from aerial images are more difficult to discern. To investigate this issue, we assess the performance of two state-of-the-art CNN algorithms, namely Faster R-CNN, which is the most popular region-based algorithm, and YOLOv3, which is known to be the fastest detection algorithm. We analyze two datasets with different characteristics to check the impact of various factors, such as UAV’s altitude, camera resolution, and object size. The objective of this work is to conduct a robust comparison between these two cutting-edge algorithms. By using a variety of metrics, we show that none of the two algorithms outperforms the other in all cases.

Objective: The aim of study is to find key genes and enriched pathways associated with lung cancer. Participants and Methods: Differentially expressed genes (DEGs) data of 54674 genes based on stage, tumor and status of lung cancer was taken from 66 patients of African American (AAs) origin. 2392 DEGs were found based on stage, 13502 DEGs were found based on tumor, 2927 DEGs were found based on status having p value (p<0.05). Results: Total 33 common DEGs were found from stage, tumor and status of lung cancer. Gene ontology (GO) and KEGG pathway enrichment analysis was performed and 49 significant pathways were obtained, out of which 10 pathways were found to be exclusively involved in lung cancer development. Protein-protein interaction (PPI) network analysis found 69 nodes and 324 edges and identified 10 hub genes based on their highest degrees. Module analysis of PPI found that ‘Viral carcinogenesis’, ‘pathways in cancer’, ‘notch signaling pathway’, ‘AMPK signaling pathways’ had a close association with lung cancer. Conclusion: These identified DEGs regulate other genes which play important role in growth of lung cancer. The key genes and enriched pathways identified can thus help in better identification and prediction of lung cancer.

The Internet of Things (IoT) is an emerging technology that aims to enable the interconnection of a large number of smart devices and heterogeneous networks. Ad hoc networks play an important role in the designing of IoT-enabled platforms due to their efficient, flexible, low-cost, and dynamic infrastructures. These networks utilize the available resources efficiently to maintain the Quality of Service (QoS) in a multi-hop communication. However, in a multi-hop communication, the relay nodes can be malicious, thus requiring a secured and reliable data transmission. In this paper, we propose a QoS-aware secured communication scheme for IoT-based networks (QoS-IoT). In QoS-IoT, a Sybil attack detection mechanism is used for the identification of Sybil nodes and their forged identities in multi-hop communication. %by high-power and mobile nodes. After Sybil nodes detection, an optimal contention window (CW) is selected for QoS provisioning, i.e., to achieve per-flow fairness and efficient utilization of the available bandwidth. In a multi-hop communication, the MAC layer protocols do not perform well in terms of fairness and throughput, especially when the nodes generate a large amount of data. It is because the MAC layer has no capability of providing QoS to prioritized or forwarding flows. We evaluate the performance of QoS-IoT in terms of Sybil attack detection, fairness, throughput, and buffer utilization. The simulation results show that the proposed scheme outperforms the existing schemes and significantly enhances the performance of the network with a large volume of data. Moreover, the proposed scheme is resilient against Sybil attack.

Opportunistic routing is the latest technique that uses the broadcasting nature of the wireless medium to increase the number of potential forwarding nodes in the network. This, in turn improves the delivery rate and reliability of data transmission in the network. Compared to all previous classes of routing protocols, opportunistic routing offers numerous advantages which is exploited by the latest applications for efficient communication and resource sharing in dynamic ad hoc networks. These applications provide dynamic communication in disaster recovery environments. The objective of this research work is to review and classify all the major opportunistic routing protocols proposed for dynamic ad hoc networks. Further the issues and challenges with each of these existing protocol is discussed and future research directions are put forward

Social media has become the most influential battleground for political and societal debate. The reach, the speed, and the fact that everybody can participate make this medium very popular but also highly manipulative, subject to biases and extremely precious to control. Misinformation spreads in an unforeseeable way. Sometimes it just follows usual non-malicious social interaction patterns, but often a targeted and designed-for-purpose manner. In this paper, we propose a novel method to identify the most effective opportunities to moderate the social media debate at a large scale using the properties of social signed graphs, such as structural balance. After building the topic universe for a particular political debate in a given time frame, we extract the most active and influential users and mine their position regarding the topic based on their interactions with other users. The result is a signed graph where each edge represents either support or challenge between a pair of users in the context of the topic. This graph can be modelled as an Ising model, with the goal of minimizing its energy state by using the advantages of Adiabatic Quantum Computing. Applying this principle, we use a Quantum Annealing implementation to solve the problem of finding the edges violating the structural balance of this graph, which in classic computing is known to be an NP-hard problem. We then provide a non-intrusive moderation technique for those violations to mitigate or increase the structural imbalance, depending on the desired strategy. The methodology is shown in the context of the COVID-19 skeptics’ debate in Twitter and implemented using a D-Wave 2000Q-6 quantum annealer.

The study presents a mathematical modeling for supporting the premise that our body is composed of blockchain systems. A cell population is considered a space with peer-to-peer communication networks for applying blockchain protocol. Transaction is defined as gene expression that constantly occurs for protein synthesis in each cell. The transaction process proceeds according to the blockchain protocol with sharing and recording the data on the blockchain ledger. The premise comes from the clues of the previous research such that the cell population is a complex structured network system having cell-to-cell connections. Although individual cells exhibit stochastic nonlinear dynamic behavior, cell population shows consensus behavior that reaches to ensemble through interaction among cells. It is inferred that gene expression is not regulated by the corresponding cell only nor determined by external intelligence such as the brain. This is achieved through a consensus through stochastic interactions between cells over the whole cell population. These findings imply that mathematical blockchain modeling is a suitable for gene expression process. The original contribution of the study is a methodology for applying mathematically the blockchain protocol to the real biological gene expression process. In other words, the DNA sequence is converted into a numeral bit sequence that makes it possible to implement the blockchain protocol. A new DNA sequence scheme is proposed with adding methylated cytosine and adenine as the 5th and 6th bases for including epigenetic information which has profound effect on gene expression and regulation. The methodology was applied to the real biological synthesis process of protein samples. The protein is composed of amino acids that are encoded by triplet codons of 216 kinds with 6 base RNA sequence. The gene expression information is traced backward from a synthesized protein sample, amino acids of codons, RNA transcript up to DNA sequence. One of the results is a numeric value in the form of a bit sequence with which mathematical blockchain modeling is applicable. The cryptographic authentication and the consensus process are mathematically proven to work properly by the blockchain protocol. It implies that the same protein is synthesized, but with different epigenetic data, then protein's latent material properties will certainly be different. Although the result has not been justified by the biologic experiment yet, it is sure that the biological hidden algorithm inside DNA sequence will be revealed by the binary bit-logic with physical on/off states which is mathematically proven. The research will greatly contribute to disease treatment and medicine development as well as epigenetics in the future.

Satellite Image Time Series (SITS) is a data set which includes satellite images across several years with a high acquisition rate. Radiometric normalization is a fundamental and important preprocessing method for remote sensing applications using SITS due to the radiometric distortion caused by noise between images. Normalizing the subject image based on the reference image is a general strategy when using traditional radiometric normalization methods to normalize multi-temporal imagery (usually two scenes or three scenes in different time phases). However, these methods are not suitable for calibrating SITS because they cannot minimize the radiometric distortion between any pair of images in SITS. The existing relative radiometric normalization methods for SITS are based on linear assumptions which cannot effectively reduce nonlinear radiometric distortion caused by continuously changing noise in SITS. To overcome this problem and obtain a more accurate SITS, this study proposed a Nonlinear Radiometric Normalization Model (named NMAG) for SITS based on Artificial Neural Networks (ANN) and Greedy Algroithm (GA) . In this method, GA was used to determine the correction order of SITS and calculate the error between the image to be corrected and normalized images, which avoided the selection of a single reference image. ANN was used to obtain the optimal solution of error function, which minimized the radiometric distortion between different images in SITS. SITS composed of 21 Landsat-8 images in Tianjin City from October 2017 to January 2019 were selected to test the method. We compared NMAG with other two contrasting method (refered as CM1 and CM2), and found that the average of root mean square error $(\mu_{RMSE})$ of NMAG $(497.22)$ was significantly smaller than those of CM1 $(641.39)$ and CM2 $(543.47)$, and the accuracy of normalized SITS obtained by using NMAG has increased by 22.4\% and 8.5\% toward CM1 and CM2, respectively. These experimental results confirmed the effectiveness of NMAG in the reduction of radiometric distortion caused by continuously changing noise between images in SITS.

Interferometry Synthetic Aperture Radar (InSAR) is an advanced remote sensing technique for studying the earth's surface topography and deformations. It is used to generate high-quality Digital Elevation Models (DEMs). DEMs are a crucial and primary input to various topographical quantification and modelling applications. The quality of input DEMs can be further improved using fusion methods, which combine multi-sensor or multi-temporal datasets intelligently to retrieve the best information amongst the input data. This research study is based on developing a Neural Network based fusion approach for improving InSAR based DEMs in plain and hilly terrains. The study areas comprise of relatively plain terrain from Ghaziabad and hilly terrain of Dehradun and their surrounding regions. The training dataset consists of DEM elevations and derived topographic attributes like slope, aspect, topographic position index (TPI), terrain ruggedness index (TRI), and vector roughness measure (VRM) in different land use land cover classes of the study areas. The spaceborne altimetry ICESat-2 ATL08 photon data is used as a reference elevation. A Feed Forward Neural Network with backpropagation algorithm is trained based on the prepared training samples. The trained model produces fused DEMs by learning the relationship between the input and target samples. This is used to predict elevations in the test areas. The accuracy of results from the models are assessed with TanDEM-X 90 m DEM. The fused DEMs show significant improvement in terms of RMSE over the input DEMs with improvement factor of 94.65 % in plain area and 82.62 % in hilly area. The study concludes that the ANN with its universal approximation property is able to significantly improve the fused DEM.

In the paper, a coordinated control methodology of single-phase (1-P) end-users switching oper-ations on the phases of an active electric distribution network (AEDN) has been proposed to ob-tain a minimum unbalance degree at the coupling common point (CCP) level with the main dis-tribution system. The phase load balancing (PLB) process considers the smart devices that switch from one phase to another phase the 1-P end-users (consumers and prosumers) to compensate for the phase load unbalance. The proposed methodology has been tested successfully in an AEDN belonging to a Romanian Distribution Network Operator (DNO) containing 114 end-users (104 consumers/10 prosumers) integrated into the Smart Metering System (SMS). The op-timal solution leads to a value of the objective function by 1.00004, represented by the unbalance factor, very close to the ideal target, 1.00. A comparative analysis was conducted considering other possible PLB cases (the consumer-level PLB and prosumer-level PLB), obtaining similar values of the UF (1.027 vs. 1.028), slightly higher than in the hybrid-level PLB. Also, the signifi-cant technical benefits were quantified through an energy-saving of 58.73% and decreasing the phase voltage unbalance rate by 91 % compared to the initial case (without PLB). These results emphasized the positive impact of the proposed coordinated control methodology on the PLB process and evidenced its effectiveness and applicability in the AEDNs.

In this article we present a deployment of 11 nodes using the three different SUN (Smart Utility Network) modulation schemes, as defined in the IEEE 802.15.4-2015 standard. The nodes were deployed in a 110.044 m2 warehouse for 99 days, and the resulting dataset contains a total of 10.710.868 measurements with RSSI (Received Signal Strength Indicator), CCA (Clear Channel Assessment) and PDR (Packet Delivery Ratio) values. The analyzed results show a high variability in average RSSI (i.e., between -82.1 dBm and -101.7 dBm) and CCA (i.e., between -111.2 dBm and -119.9 dBm) values, which are caused by the effects of multi-path propagation and external interference. Despite being above the sensitivity limit for each modulation, this values result in poor average PDR values (i.e., from 65.9% to 87.4%), indicating that additional schemes are required for low-power wireless communications to meet the dependability requirements of industrial applications. For that purpose, we also introduce the concept of modulation diversity, which can be combined with packet repetition to meet such requirements (i.e., PDR>99%) while minimizing the energy expenditure of nodes and meeting regulatory constraints.

Convolutional neural networks (CNNs), inspired by biological visual cortex systems, are a powerful category of artificial neural networks that can extract the hierarchical features of raw data to greatly reduce the network parametric complexity and enhance the predicting accuracy. They are of significant interest for machine learning tasks such as computer vision, speech recognition, playing board games and medical diagnosis [1-7]. Optical neural networks offer the promise of dramatically accelerating computing speed to overcome the inherent bandwidth bottleneck of electronics. Here, we demonstrate a universal optical vector convolutional accelerator operating beyond 10 Tera-OPS (TOPS - operations per second), generating convolutions of images of 250,000 pixels with 8-bit resolution for 10 kernels simultaneously — enough for facial image recognition. We then use the same hardware to sequentially form a deep optical CNN with ten output neurons, achieving successful recognition of full 10 digits with 900 pixel handwritten digit images with 88% accuracy. Our results are based on simultaneously interleaving temporal, wavelength and spatial dimensions enabled by an integrated microcomb source. This approach is scalable and trainable to much more complex networks for demanding applications such as unmanned vehicle and real-time video recognition.

Automated skin lesion analysis is one of the trending fields that has gained attention among the dermatologists and healthcare practitioners. Skin lesion restoration is an essential preprocessing step for lesion enhancements for accurate automated analysis and diagnosis. Digital hair removal is a non-invasive method for image enhancement by solving the hair-occlusion artefact in previously captured images. Several hair removal methods were proposed for skin delineation and removal. However, manual annotation is one of the main challenges that hinder the validation of these proposed methods on a large number of images or using benchmarking datasets for comparison purposes. In the presented work, we propose a realistic hair simulator based on context-aware image synthesis using image-to-image translation techniques via conditional adversarial generative networks for generation of different hair occlusions in skin images, along with the ground-truth mask for hair location. Besides, we explored using three loss functions including L1-norm, L2-norm and structural similarity index (SSIM) to maximise the synthesis quality. For quantitatively evaluate the realism of image synthesis, the t-SNE feature mapping and Bland-Altman test are employed as objective metrics. Experimental results show the superior performance of our proposed method compared to previous methods for hair synthesis with plausible colours and preserving the integrity of the lesion texture.

This study proposes a mobile positioning method which adopts recurrent neural network algorithms to analyze the received signal strength indications from heterogeneous networks (e.g., cellular networks and Wi-Fi networks) for estimating the locations of mobile statioThis study proposes a mobile positioning method which adopts recurrent neural network algorithms to analyze the received signal strength indications from heterogeneous networks (e.g., cellular networks and Wi-Fi networks) for estimating the locations of mobile stations. The recurrent neural networks with multiple consecutive timestamps can be applied to extract the features of time series data for the improvement of location estimation. In practical experimental environments, there are 4,525 records, 59 different base stations, and 582 different Wi-Fi access points detected in Fuzhou University in China. The lower location errors can be obtained by the recurrent neural networks with multiple consecutive timestamps (e.g., 2 timestamps and 3 timestamps); the experimental results can be observed that the average error of location estimation was 9.19 meters by the proposed mobile positioning method with 2 timestamps.

Wireless Sensor Networks (WSN) is the network of the resource-constrained network which forms the foundation of the Internet of Things (IoT). The Routing Protocol for Low-power and Lossy Networks (RPL) is responsible for generating and managing data routing paths. Nodes implementing RPL uses the mechanics of Objective Function (OF) to select the preferred next-hop node – parent node, and optimal routing path to the destination node. If routing decisions are not efficiently made, this results in increased collision domain, leading to packet losses and packet retransmission which impairs the network operational lifetime. In this study, we present the Contiki Operating System (OS), a state-of-the-art OS for IoTs, ContikiRPL; Contiki variant of RPL. We investigated the performance of RPL with respect to its two OFs; Objective Function Zero (OF0) and the Minimum Rank with Hysteresis Objective Function (MRHOF). The performance of these OFs was evaluated on the following metrics; Packet Delivery Ratio (PDR), Power consumption, and network latency. The result shows that MRHOF outperformed OF0 on all metrics with an overall average PDR of 91.5%, a latency of 44ms, and power consumption of 1.72mW across all nodes. This results in optimal network performance with improved network operational lifetime.

In the paper, an improved smart meter data-based three-stage algorithm to calculate the power/energy losses in the three-phase networks with the voltage level below 0.4 kV (low voltage - LV) is presented. In the first stage, a loading function of input data was built having as main feature the working at the same time with files from the database of smart metering system (SMS) containing the hourly electricity records, and files including the characteristic load profiles established by the Distribution Network Operator (DNO) for the consumers with standard energy meters depending the following factors: consumption class, day and season. In the second stage, a function which is based on the work with the structure vectors was implemented to identify easy the configuration of analysed networks. In the third stage, an improved version of forward/backward sweep-based algorithm was proposed to calculate fast the power/energy losses to three-phase LV distribution networks in balanced and unbalanced regime. A real LV rural distribution network from a pilot zone belonging to a Distribution Network Operator (DNO) from Romania was used to confirm the accuracy of the proposed approach. The comparison with the results obtained using the DigSilent PowerFactory Simulation Package certified the performance of the algorithm, the mean absolute percentage error (MAPE) being 0.94%.

1). As on-the-ground projects come into existence and continue to expand to adapt to climate change and empower women, it is important to understand their location within TANs. Using the Bhungroo technology as a case study, we aim to assess the potential of TANs to increase the scope and scale of local projects as well as the ability of similar and emerging projects to create social change at local levels. 2). We do so by analyzing interview and earned media hits data from the UNFCCC Momentum for Change. 3). We find that while TANs may help increase the scale and scope of projects, increasing their ability to effectively reach more people and areas is still up for debate 4). We conclude by considering how women’s political participation may be enhanced by similar projects.

Diabetic retinopathy is the most common microvascular complication of diabetes mellitus and one of the leading causes of blindness globally. Due to the progressive nature of the disease, earlier detection and timely treatment can lead to substantial reductions in the incidence of irreversible vision-loss. Artificial intelligence (AI) screening systems have offered clinically acceptable and quicker results in detecting diabetic retinopathy from retinal fundus and optical coherence tomography (OCT) images. Thus, this systematic review and meta-analysis of relevant investigations was performed to document the performance of AI screening systems that were applied to fundus and OCT images of patients from diverse geographic locations including North America, Europe, Africa, Asia, and Australia. A systematic literature search on Medline, Global Health, and PubMed was performed and studies published between October 2015 and January 2020 were included. The search strategy was based on the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting guidelines, and AI-based investigations were mandatory for studies inclusion. The abstracts, titles, and full-texts of potentially eligible studies were screened against inclusion and exclusion criteria. Twenty-one studies were included in this systematic review; 18 met inclusion criteria for the meta-analysis. The pooled sensitivity of the evaluated AI screening systems in detecting diabetic retinopathy was 0.93 (95% CI: 0.92-0.94) and the specificity was 0.88 (95% CI: 0.86-0.89). The included studies detailed training and external validation datasets, criteria for diabetic retinopathy case ascertainment, imaging modalities, DR-grading scales, and compared AI results to those of human graders (e.g., ophthalmologists, retinal specialists, trained nurses, and other healthcare providers) as a reference standard. The findings of this study showed that the majority AI screening systems demonstrated clinically acceptable levels of sensitivity and specificity for detecting referable diabetic retinopathy from retinal fundus and OCT photographs. Further improvement depends on the continual development of novel algorithms with large and gradable sets of images for training and validation. If cost-effectiveness ratios can be optimized, AI can become a financially sustainable and clinically effective intervention that can be incorporated into the healthcare systems of low-to-middle income countries (LMICs) and geographically remote locations. Combining screening technologies with treatment interventions such as anti-VEGF therapy, acellular capillary laser treatment, and vitreoretinal surgery can lead to substantial reductions in the incidence of irreversible vision-loss due to proliferative diabetic retinopathy.

As time progresses and technology improves, biological data sets are continuously increasing in size. New methods and new implementations of existing methods are needed to keep pace with this increase. In this paper, we present a high performance computing(HPC)-capable implementation of Iterative Random Forest (iRF). This new implementation enables the explainable-AI eQTL analysis of SNP sets with over a million SNPs. Using this implementation we also present a new method, iRF Leave One Out Prediction (iRF-LOOP), for the creation of Predictive Expression Networks on the order of 40,000 genes or more. We compare the new implementation of iRF with the previous R version and analyze its time to completion on two of the world's fastest supercomputers Summit and Titan. We also show iRF-LOOP's ability to capture biologically significant results when creating Predictive Expression Networks. This new implementation of iRF will enable the analysis of biological data sets at scales that were previously not possible.

Link prediction is an iconic problem in complex networks because deals with the ability to predict nonobserved existing or future parts of the network structure. The impact of this prediction on real applications can be disruptive: from prediction of covert links between terrorists in their social networks to repositioning of drugs in molecular diseasome networks. Here we compare: (1) an ensemble meta-learning method (Ghasemian et al.), which uses an artificial intelligence (AI) stacking strategy to create a single meta-model from hundreds of other models; (2) a structural predictability method (SPM, Lü et al.), which relies on a theory derived from quantum mechanics and does not assume any model; (3) a modelling rule named Cannistraci-Hebb (CH, Muscoloni et al.), which relies on one brain-bioinspired model adapting to the intrinsic network structure.We conclude that brute-force stacking of algorithms by AI does not perform better than (and is often significantly outperformed by) SPM and one simple brain-bioinspired rule such as CH. This agrees with the Gödel incompleteness: stacking is optimal but incomplete, you cannot squeeze out more than what is already in your features. Hence, we should also pursue AI that resembles human-like physical ‘understanding’ of simple generalized rules associated to complexity. The future might be populated by AI that ‘steals for us the fire from Gods’, towards machine intelligence that creates new rules rather than stacking the ones already known.

Corynebacterium pseudotuberculosis is a Gram-positive bacterium that causes caseous lymphadenitis, a disease that predominantly affects sheep, goat, cattle, buffalo, and horses, but has also been recognized in other animals. This bacterium generates a severe economic impact on countries producing meat. Gene expression studies using RNA-seq is one of the most commonly used techniques to perform transcriptional experiments. Computational analysis on such data through reverse-engineering algorithms leads to a better understanding of the genome-wide complexity of gene interactomes, enabling the identification of genes having the most significant functions inferred by the activated stress response pathways. In this study, we identified the influential or causal genes from four RNA-seq data-sets from different stress conditions (high iron, low iron, acid, osmosis, and PH) in C. pseudotuberculosis, using a consensus-based network inference algorithm called miRsig and identified the causal genes in the network using the miRinfluence tool, which is based on the influence diffusion model. We found that over 50\% of the genes identified as influential have some essential cellular functions in the genomes. In the strains analyzed, most of the causal genes have crucial roles or participate in processes associated with response to extracellular stresses, pathogenicity, membrane components, and essential genes. This research brings new insight into the understanding of virulence and infection by C. pseudotuberculosis.

The aim of this expository article is to present recent developments in the centuries-old discussion on the interrelations between several types of domination in graphs. However, the novelty even more prominent in the newly discovered simplified presentations of several older results. Domination can be seen as arising from real-world application and extracting classical results as first described by this article.The main part of this article, concerning a new domination and older one, is presented in a narrative that answers two classical questions: (i) To what extend must closing set be dominating? (ii) How strong is the assumption of domination of a closing set? In a addition, we give an overview of the results concerning domination. The problem asks how small can a subset of vertices be and contain no edges or, more generally how can small a subset of vertices be and contain other ones. Our work was as elegant as it was unexpected being a departure from the tried and true methods of this theory that had dominated the field for one fifth a century. This expository article covers all previous definitions. The inability of previous definitions in solving even one case of real-world problems due to the lack of simultaneous attentions to the worthy both of vertices and edges causing us to make the new one. The concept of domination in a variety of graphs models such as crisp, weighted and fuzzy, has been in a spotlight. We turn our attention to sets of vertices in a fuzzy graph that are so close to all vertices, in a variety of ways, and study minimum such sets and their cardinality. A natural way to introduce and motivate our subject is to view it as a real-world problem. In its most elementary form, we consider the problem of reducing waste of time in transport planning. Our goal here is to first describe the previous definitions and the results, and then to provide an overview of the flows ideas in their articles. The final outcome of this article is twofold: (i) Solving the problem of reducing waste of time in transport planning at static state; (ii) Solving and having a gentle discussions on problem of reducing waste of time in transport planning at dynamic state. Finally, we discuss the results concerning holding domination that are independent of fuzzy graphs. We close with a list of currently open problems related to this subject. Most of our exposition assumes only familiarity with basic linear algebra, polynomials, fuzzy graph theory and graph theory.

It's challenging to achieve robust lane detection depending on single frame when considering complicated scenarios. In order to detect more credible lane markings by using sequential frames, a novel approach to fusing vision and Inertial Measurement Unit (IMU) is proposed in this paper. The hough space is employed as the space where lane markings are stored and it's calculated by three steps. Firstly, a basic hough space is extracted by Hough Transform and primary line segments are extracted from it. In order to measure the possibility about line segments belong to lane markings, a CNNs based classifier is introduced to transform the basic hough space into a probabilistic space by using the networks outputs. However, this probabilistic hough space based on single frame is easily disturbed. In the third step, a filtering process is employed to smooth the probabilistic hough space by using sequential information. Pose information provided by IMU is applied to align hough spaces extracted at different times to each other. The final hough space is used to eliminate line segments with low possibility and output those with high confidence as the result. Experiments demonstrate that the proposed approach has achieved a good performance.

Deep learning classification is the state-of-the-art of machine learning approach. Earlier work proves that the deep convolutional neural network has successfully and brilliantly in different applications such as images or video data. Recognizing and clarifying the remote sensing aspect of the earth's surface and exploit land cover and land use (LCLU). First, this article summarized the remote sensing emerging application and challenges for deep learning methods. Second, we propose four approaches to learn efficient and effective CNNs to transfer image representation on the ImageNet dataset to recognize LCLU datasets. We use VGG16, Inception-ResNet-V2, Inception-V3, and DenseNet201 models to extract features from the EACC dataset. We use pre-trained CNNs on ImageNet to extract features. For feature selection we proposed principal component analysis (PCA) to improve accuracy and speed up the model. We train our model by multi-layer perceptron (MLP) as a classifier. Lastly, we apply the multi-granularity encoding ensemble model. We achieve an overall accuracy of 92.3% for the nine-class classification problem. This work will help remote sensing scientists understand deep learning tools and apply them in large-scale remote sensing challenges

This work presents an approach for detecting mathematical formulas in scanned document images. The proposed approach is end-to-end trainable. Since many OCR engines cannot reliably work with the formulas, it is essential to isolate them to obtain the clean text for information extraction from the document. Our proposed pipeline comprises a hybrid task cascade network with deformable convolutions and a Resnext101 backbone. Both of these modifications help in better detection. We evaluate the proposed approaches on the ICDAR-2017 POD and Marmot datasets and achieve an overall accuracy of 96% for the ICDAR-2017 POD dataset. We achieve an overall reduction of error of 13%. Furthermore, the results on Marmot datasets are improved for the isolated and embedded formulas. We achieved an accuracy of 98.78% for the isolated formula and 90.21% overall accuracy for embedded formulas. Consequently, it results in an error reduction rate of 43% for isolated and 17.9% for embedded formulas.

Hopfield Neural Networks (HNNs) are recurrent neural networks used to implement associative memory. Their main feature is their ability to pattern recognition, optimization, or image segmentation. However, sometimes it is not easy to provide the users with good explanations about the results obtained with them due to mainly the large number of changes in the state of neurons (and their weights) produced during a problem of machine learning. There are currently limited techniques to visualize, verbalize, or abstract HNNs. This paper outlines how we can construct automatic video generation systems to explain their execution. This work constitutes a novel approach to get explainable artificial intelligence systems in general and HNNs in particular building on the theory of data-to-text systems and software visualization approaches. We present a complete methodology to build these kinds of systems. Software architecture is also designed, implemented, and tested. Technical details about the implementation are also detailed and explained. Finally, we apply our approach for creating a complete explainer video about the execution of HNNs on a small recognition problem.

Data sets for medical images are generally imbalanced and limited in sample size because of high data collection costs, time-consuming annotations, and patient privacy concerns. The training of deep neural network classification models on these data sets to improve the generalization ability does not produce the desired results for classifying the medical condition accurately and often overfit the data on the majority of class samples. To address the issue, we propose a framework for improving the classification performance metrics of deep neural network classification models using transfer learning: pre-trained models, such as Xception, InceptionResNet, DenseNet along with the Generative Adversarial Network (GAN) – based data augmentation. Then, we trained the network by combining traditional data augmentation techniques, such as randomly flipping the image left to right and GAN-based data augmentation, and then fine-tuned the hyper-parameters of the transfer learning models, such as the learning rate, batch size, and the number of epochs. With these configurations, the Xception model outperformed all other pre-trained models achieving a test accuracy of 98.7%, the precision of 99%, recall of 99.3%, f1-score of 99.1%, receiver operating characteristic (ROC) - area under the curve (AUC) of 98.2%.

This paper reviews non-intrusive load monitoring (NILM) approaches that employ deep neural networks to disaggregate appliances from low frequency data, i.e. data with sampling rates lower than the AC base frequency. We first review the many degrees of freedom of these approaches, what has already been done in literature, and compile the main characteristics of the reviewed publications in an extensive overview table. The second part of the paper discusses selected aspects of the literature and corresponding research gaps. In particular, we do a performance comparison with respect to reported MAE and F$_1$-scores and observe different recurring elements in the best performing approaches, namely data sampling intervals below 10\,s, a large field of view, the usage of GAN losses, multi-task learning, and post-processing. Subsequently, multiple input features, multi-task learning and related research gaps are discussed, the need for comparative studies is highlighted, and finally, missing elements for a successful deployment of NILM approaches based on deep neural networks are pointed out. We conclude the review with an outlook on possible future scenarios.

Future cyber-physical systems may extend over broad geographical areas, like cities or regions, thus requiring the deployment of large real-time networks. A strategy to guarantee predictable communication over such networks is to synthesize an offline time-triggered communication schedule. However, this synthesis problem is computationally hard (NP-complete), and existing approaches do not scale satisfactorily to the required network sizes. This article presents a segmented offline synthesis method which substantially reduces this limitation, being able to generate time-triggered schedules for large hybrid (wired and wireless) networks. We also present a series of algorithms and optimizations that increase the performance and compactness of the obtained schedules while solving some of the problems inherent to segmented approaches. We evaluate our approach on a set of realistic large-size multi-hop networks, significantly larger than those considered in the existing literature. The results show that our segmentation reduces the synthesis time up to two orders of magnitude.

The EigenAnt algorithm has been proposed to find the shortest path between two nodes with a proof of convergence. Two novel features are noticeable in EigenAnt compared to the conventional ant colony optimization algorithms. Pheromone evaporation in the EigenAnt algorithm is done locally for the selected path and it does not use heuristic information in its selection phase. On the other hand, a simple version of ant colony optimization that also does not use heuristic information has been used for analyzing convergence, considering parameter impact analysis, in the problem of finding the shortest path for 1-node binary chain problems. In this paper, we propose to improve the EigenAnt algorithm by adding additional parameters in such a way as to be able to take advantage of the analysis developed for the simple ant colony optimization algorithm. We demonstrate through our analysis that since the Improved EigenAnt algorithm has one parameter more than ant colony optimization algorithms, tuning of convergence speed is decoupled from tuning of convergence features in the algorithm. Thus, IEigenAnt, which can be interpreted as having two uncorrelated intensification and diversification components, makes it possible to achieve a balance between intensification and diversification. In order to illustrate the advantages of this decoupling, we present experimental results for the routing network problem.

Precipitation downscaling is widely employed for enhancing the resolution and accuracy of precipitation products from general circulation models (GCMs). In this study, we propose a novel statistical downscaling method to foster GCMs’ precipitation prediction resolution and accuracy for monsoon region. We develop a deep neural network composed of convolution and Long Short Term Memory (LSTM) recurrent module to estimate precipitation based on well-resolved atmospheric dynamical fields. The proposed model is compared against GCM precipitation product and classical downscaling methods in the Xiangjiang River Basin in South China. Results show considerable improvement compared to the ECMWF-Interim reanalysis precipitation. Also, the model outperforms benchmark downscaling approaches, including 1) quantile mapping, 2) support vector machine, and 3) convolutional neural network. To test the robustness of the model and its applicability in practical forecast, we apply the trained network for precipitation prediction forced by retrospective forecasts from ECMWF model. Compared to ECMWF precipitation forecast, our model makes better use of the resolved dynamical field for more accurate precipitation prediction at lead time from 1 day up to 2 weeks. This superiority decreases along forecast lead time, as GCM’s skill in predicting atmospheric dynamics being diminished by the chaotic effect. At last, we build a distributed hydrological model and force it with different sources of precipitation inputs. Hydrological simulation forced with the neural network precipitation estimation shows significant advantage over simulation forced with the original ERA-Interim precipitation (with NSE value increases from 0.06 to 0.64), and the performance is just slightly worse than the observed precipitation forced simulation (NSE=0.82). This further proves the value of the proposed downscaling method, and suggests its potential for hydrological forecasts.

Deep neural networks (DNNs), founded on the brain's neuronal organization, can extract higher-level features from raw input. However, complex intellect via autonomous decision-making is way beyond current AI design. Here we propose an autonomous AI inspired by the thermodynamic cycle of sensory perception, operating between two information density reservoirs. Stimulus unbalances the high entropy resting state and triggers a thermodynamic cycle. By recovering the initial conditions, self-regulation generates a response while accumulating an orthogonal, holographic potential. The resulting high-density manifold is a stable memory and experience field, which increases future freedom of action via intelligent decision-making.

In this paper, we examine two strategies for boosting the performance of ensembles of Siamese networks (SNNs) for image classification using two loss functions (Triplet and Binary Cross Entropy) and two methods for building the dissimilarity spaces (FULLY and DEEPER). With FULLY, the distance between a pattern and a prototype is calculated by comparing two images using the fully connected layer of the Siamese network. With DEEPER, each pattern is described using a deeper layer combined with dimensionality reduction. The basic design of the SNNs takes advantage of supervised k-means clustering for building the dissimilarity spaces that train a set of support vector machines, which are then combined by sum rule for a final decision. The robustness and versatility of this approach are demonstrated on several cross-domain image data sets, including a portrait data set, two bioimage and two animal vocalization data sets. Results show that the strategies employed in this work to increase the performance of dissimilarity image classification using SNN is closing the gap with standalone CNNs. Moreover, when our best system is combined with an ensemble of CNNs, the resulting performance is superior to an ensemble of CNNs, demonstrating that our new strategy is extracting additional information.

This study introduces a comparative analysis of two deep learning (multilayer perceptron neural networks (MLP-NN) and the long short term memory networks (LSTMN)) models for transit travel time prediction. The two models were trained and tested using one-year worth of data for a bus route in Blacksburg, Virginia. In this study, the travel time was predicted between each two successive stations to all the model to be extended to include bus dwell times. Additionally, two additional models were developed for each category (MLP of LSTM): one for only segments including controlled intersections (controlled segments) and another for segments with no control devices along them (uncontrolled segments). The results show that the LSTM models outperform the MLP models with a RMSE of 17.69 sec compared to 18.81 sec. When splitting the data into controlled and uncontrolled segments, the RMSE values reduced to 17.33 sec for the controlled segments and 4.28 sec for the uncontrolled segments when applying the LSTM model. Whereas, the RMSE values were 19.39 sec for the controlled segments and 4.67 sec for the uncontrolled segments when applying the MLP model. These results demonstrate that the uncertainty in traffic conditions introduced by traffic control devices has a significant impact on travel time predictions. Nonetheless, the results demonstrate that the LSTMN is a promising tool that can has the ability to account for the temporal correlation within the data. The developed models are also promising tools for reasonable travel time predictions in transit applications.

The objective of this article was to forecast the ultimate failure load laminate stacking sequence combination on bonding joints which are exposed to tensile strength by using artificial neural networks. We have glass fiber composite materials with three different sequence combinations ([0°/90°], [±45°], [0°/90°/±45°]). Various adherend thicknesses and also ductile type adhesive was used in the experiment. The bonding geometry is a single lap and has four types of overlap angles 30°, 45°, 60°, 75° respectively. The experimental results demonstrate that composite laminate stacking sequence profoundly affects the bonding joints of failure load. Taking experimental results into account, Levenberg–Marquardt learning algorithm model was used by preferring a three layer forward on ANN so as to discipline network. In order to procure a precise ANN tool, an integrate methodology of experimental method has been used. The outcomes are used to ensure the experimental data’s to the ANN. The method of ANN permits surveying much adequately the probabilities of composite laminate stacking sequence combination using the prevalent ones which are [0°/90°], [±45°] and [0°/90°/±45°]. Testing data and training results were quite well 0.998, 0.997 and 0.998 in turn. Consequences acquired can be used by engineers who are interested in the composite material design to enhance failure load.

This review covers some of the central relationships between artificial intelligence, psychology, and economics through the lens of information theory, specifically focusing on formal models of decision-theory. In doing so we look at a particular approach that each field has adopted and how information theory has informed the development of the ideas of each field. A key theme is expected utility theory, its connection to information theory, and the Bayesian approach to decision-making and forms of (bounded) rationality. What emerges from this review is a broadly unified formal perspective derived from three very different starting points that reflect the unique principles of each field. Each of the three approaches reviewed can, in principle at least, be implemented in a computational model in such a way that, with sufficient computational power, they could be compared with human abilities in complex tasks. However, a central critique that can be applied to all three approaches was first put forward by Savage in {\it The Foundations of Statistics} and recently brought to the fore by the economist Binmore: Bayesian approaches to decision-making work in what Savage called `small worlds' but cannot work in `large worlds'. This point, in various different guises, is central to some of the current debates about the power of artificial intelligence and its relationship to human-like learning and decision-making. Recent work on artificial intelligence has gone some way to bridging this gap but significant questions still need to be answered in all three fields in order to make progress on these problems.

In this paper, we propose and evaluate two mechanisms aimed at improving the communication reliability of IEEE 802.15g SUN (Smart Utility Networks) in industrial scenarios: RTS (Re-Transmission Shaping), which uses acknowledgements to track channel conditions and dynamically adapt the number of re-transmissions per packet, and AMS (Adaptive Modulation Selection), which makes use of reinforcement learning based on MAB (Multi-Armed Bandits) to choose the modulation that provides the best reliability for each packet re-transmission. The evaluation of both mechanisms is performed through computer simulations using a dataset obtained from a real-world deployment and two widely used metrics, the PDR (Packet Delivery Ratio) and the RNP (Required Number of Packet transmissions). The PDR measures the ratio between received and transmitted packets, whereas the RNP is the number of packet repetitions before a successful transmission. The results show that both mechanisms allow to increase the communication reliability while not jeopardizing the battery life-time constraints of end devices. For example, when three re-transmissions per packet are allowed, the PDR reaches 98/96\% with a RNP of 2.03/1.32 using RTS and AMS, respectively. Additionally, the combination of both proposed mechanisms allows to reach a 99% PDR with a RNP of 1.7, making IEEE 802.15.4g SUN compliant with the stringent data delivery requirements of industrial applications.

The generalization of Farey graphs and extended Farey graphs are all originated from Farey graph and scale-free and small-world simultaneously. We propose a labeling of the vertices for it that allows determining all the shortest paths routing between any two vertices only based on their labels. The maximum number of shortest paths between any two vertices is huge as the product of two Fibonacci numbers, however, the label-based routing algorithm runs in linear time O(n). The existence of an efficient routing protocol for Farey-type models should help the understanding of several physical dynamic processes on it.

Automated lip reading is a research problem that has developed considerably in recent years. Lip reading is evaluated both visually and audibly in some cases. The lip reading model is a field of use for detecting specific words using images from security cameras, but it is not possible to use audio-visual databases in this situation. It is not possible to obtain the sound input of the pronounced word in all cases. We collected a new Turkish dataset with only the image in this study. The new dataset is produced using Youtube videos, which is an uncontrolled environment. For this reason, images have difficult parameters in terms of environmental factors such as light, angle, color, and personal characteristics of the face. Despite the different features on the human face such as mustache, beard, and make-up, the visual speech recognition problem was developed on 10 classes including single words and two-word phrases using Convolutional Neural Networks (CNN) without any intervention on the data. The proposed study using only-visual data obtained a model which is automated visual speech recognition with a deep learning approach. In addition, since this study uses only-visual data, the computational cost and resource usage is less than in multi-modal studies. It is also the first known study to address the lip reading problem with a deep learning algorithm using a new dataset belonging to the Ural-Altaic languages.

Teeth segmentation plays a pivotal role in dental diagnosis, treatment, planning, and the development of computer-aided dental systems. It enables precise identification and analysis of dental structures, aiding in detecting dental abnormalities, measuring tooth dimensions, and assessing oral health conditions. Accurate teeth segmentation also facilitates the automation of dental workflows, leading to improved efficiency and reduced human error. Artificial Intelligence (AI) has witnessed rapid advancements, with various approaches developed or enhanced. While Convolutional Neural Networks (CNNs) have been widely used in medical image segmentation, the U-Net architecture has emerged as a standout performer due to its exceptional segmentation capabilities. This paper presents a proof of concept for the Attention U-Net architecture, as previously used in [1], applied to teeth segmentation. The study demonstrates the superior performance of this network in accurately segmenting teeth using a newly available benchmark dataset called Tufts Dental X-Ray Dataset. When trained and tested on 10-fold cross-validation, the model achieved an average dice coefficient of 95.01%, intersection over union of 90.6%, and pixel accuracy of 98.82%. These scores surpass those of all other networks implemented on the same dataset. By leveraging the Attention U-Net architecture, our research showcases the potential of advanced AI techniques in dental radiography. The findings contribute to the ongoing efforts to develop automated systems to assist dental professionals in their clinical practice.

The upgrade of energy infrastructure to those of smart grids, necessarily goes through the provision of integrated technological solutions that ensure the interoperability of their business capabilities and reduce the risk of devaluation of the systems used. The heterogeneity of the infrastructures and the dynamics of their operating environment, requires the continuous reduction of the complexity, the faster execution of the processes and the easy addition of innovative counterparts. Also, the integrated management of the overall ecosystem demands the provision of end-to-end interconnection, quality assurance, the definition of strict security policies, collaborative integration and correlation of events. In this respect, every design detail can be critical to the success or failure of a costly and ambitious project, such as that of smart energy networks. This work presents communication operating standards specific to the smart electricity networks applications, which should be taken into account in the process of planning and implementation of new infrastructures.

The postnatal rodent spinal cord in-vitro is a useful model to investigate early pathophysiological changes after injury. While low dose nicotine (1µM) induces neuroprotection, how higher doses affect spinal networks is unknown. Using spinal preparations of postnatal wild-type Wistar rat and Wnt1Cre2:Rosa26Tom double-transgenic mouse, we studied the effect of nicotine (0.5-10µM) on locomotor networks in-vitro. Nicotine 10µM induced motoneuron depolarization, suppressed monosynaptic reflexes, and decreased fictive locomotion in rat spinal cord. Delayed fall in neuronal numbers (including motoneurons) of central and ventral regions emerged without loss of dorsal neurons. Conversely, nicotine (0.5-1µM) preserved neurons throughout the spinal cord and strongly activated the Wnt1 signaling pathway. High-dose nicotine enhanced expression of S100 and GFAP in astrocytes suggesting their response to stress. Excitotoxicity induced by kainate was contrasted by nicotine (10µM) in the dorsal area and persisted in central and ventral regions with no change basal Wnt signaling. When combining nicotine with kainate, the activation of Wnt1 was reduced compared to kainate/sham. The present results suggest that high dose nicotine was neurotoxic to central and ventral spinal neurons as the neuroprotective role of Wnt signaling became attenuated. This also corroborates the risk of cigarette smoking for the foetus/newborn since tobacco contains nicotine.

With the widespread of embedded sensing capabilities of mobile devices, there has been unprecedented development of context-aware solutions. This allows the proliferation of various intelligent applications such as those for remote health and lifestyle monitoring, intelligent personalized services, etc. However, activity context recognition based on multivariate time series signals obtained from mobile devices in unconstrained conditions is naturally prone to imbalance class problems. This means that recognition models tend to predict classes with the majority number of samples whilst ignoring classes with the least number of samples, resulting in poor generalization. To address this problem, we propose to augment the time series signals from inertia sensors with signals from ambient sensing to train deep convolutional neural networks (DCNN) models. DCNN provides the characteristics that capture local dependency and scale invariance of these combined sensor signals. Consequently, we developed a DCNN model using only inertial sensor signals and then developed another model that combined signals from both inertia and ambient sensors aiming to investigate the class imbalance problem by improving the performance of the recognition model. Evaluation and analysis of the proposed system using data with imbalanced classes show that the system achieved better recognition accuracy when data from inertial sensors are combined with those from ambient sensors such as environment noise level and illumination, with an overall improvement of 5.3% accuracy.

3D reconstruction from a single 2D input is a classic problem in the field of computer vision. With the advancements in deep learning, the performance of 3D reconstruction has also significantly improved. The reconstruction task is more difficult for objects with no textures or complex deformations. This paper serves as a review of recent literature on 3D reconstruction from a single view, with a focus on deep learning methods from 2018 to 2021. Due to lack of standard datasets or 3D shape representation methods, it is hard make direct comparisons between all reviewed methods. However, this paper reviews different approaches for reconstructing 3d shape as depth maps, surface normals, point clouds and meshes; along with various loss functions and evaluation metrics used to train and evaluate these methods.

As a relatively high percentage of adenoma polyps are missed, a computer-aided diagnosis (CAD) tool based on deep learning can aid the endoscopist in diagnosing colorectal polyps or colorectal cancer in order to decrease polyps missing rate and prevent colorectal cancer mortality. Convolutional Neural Network (CNN) is a deep learning method and has achieved better results in detecting and segmenting specific objects in images in the last decade than conventional models such as regression, support vector machines or artificial neural networks. In recent years, based on the studies in medical imaging criteria, CNN models have acquired promising results in detecting masses and lesions in various body organs, including colorectal polyps. In this review, the structure and architecture of CNN models and how colonoscopy images are processed as input and converted to the output are explained in detail. In most primary studies conducted in the colorectal polyp detection and classification field, the CNN model has been regarded as a black box since the calculations performed at different layers in the model training process have not been clarified precisely. Furthermore, I discuss the differences between the CNN and conventional models, inspect how to train the CNN model for diagnosing colorectal polyps or cancer, and evaluate model performance after the training process.

Tables in the document image are one of the most important entities since they contain crucial information. Therefore, accurate table detection can significantly improve information extraction from tables. In this work, we present a novel end-to-end trainable pipeline, HybridTabNet, for table detection in scanned document images. Our two-stage table detector uses the ResNeXt-101 backbone for feature extraction and Hybrid Task Cascade (HTC) to localize the tables in scanned document images. Moreover, we replace conventional convolutions with deformable convolutions in the backbone network. This enables our network to detect tables of arbitrary layouts precisely. We evaluate our approach comprehensively on ICDAR-13, ICDAR-17 POD, ICDAR-19, TableBank, Marmot, and UNLV. Apart from the ICDAR-17 POD dataset, our proposed HybridTabNet outperforms earlier state-of-the-art results without depending on pre and post-processing steps. Furthermore, to investigate how the proposed method generalizes unseen data, we conduct an exhaustive leave-one-out-evaluation. In comparison to prior state-of-the-art results, our method reduces the relative error by 27.57% on ICDAR-2019-TrackA-Modern, 42.64% on TableBank (Latex), 41.33% on TableBank (Word), 55.73% on TableBank (Latex + Word), 10% on Marmot, and 9.67% on UNLV dataset. The achieved results reflect the superior performance of the proposed method.

Online Social Networks (OSNs), such as Facebook and Twitter, have become a very important part of many people’s daily lives. Unfortunately, the high popularity of these platforms makes them very attractive to spammers. Machine-learning (ML) techniques have been widely used as a tool to address many cybersecurity application problems (such as spam and malware detection). However, most of the proposed approaches do not consider the presence of adversaries that target the defense mechanism itself. Adversaries can launch sophisticated attacks to undermine deployed spam detectors either during training or the prediction (test) phase. Not considering these adversarial activities at the design stage makes OSNs’ spam detectors prone to a range of adversarial attacks. This paper thus surveys the attacks against Twitter spam detectors in an adversarial environment. In addition, a general taxonomy of potential adversarial attacks is proposed by applying common frameworks from the literature. Examples of adversarial activities on Twitter were provided after observing Arabic trending hashtags. A new type of spam tweet (Adversarial spam tweet), which can be used to undermine deployed classifier, were found. In addition, possible countermeasures that could increase the robustness of Twitter spam detectors against such attacks are investigated.