Traditional Network Model Research Articles

Abnormal functional connectivity of brain regions associated with fear network model in panic disorder

Background: Patients with panic disorder (PD) have an abnormal function in brain regions related to fear network is well recognised. However, the traditional fear network model (FNM) which was based on animals’ horrible behaviours has been found that it’s not enough to explain the pathological mechanism of PD. This study aims to explore brain regions’ abnormalities in the new advanced FNM, and estimate whether it can better explain PD. Methods: Magnetic resonance imaging resting-state scans were acquired in 40 patients with PD (35 drug-naïve and 5 drug-free) and 40 healthy controls (HCs). Twelve brain regions in the advanced FNM were chosen as regions of interest (ROIs) to examine the group difference in the ROIs and whole-brain resting-state functional connectivity (rsFC). Results: We found significantly increased thalamic rsFC with the insula, compared with HCs. And it was significantly correlated with HAMA-somatic score. We also found increased thalamic rsFC with occipital gyrus, temporal gyrus, and frontal gyrus when compared with HCs. Conclusions: Taken together, PD patients exhibit abnormal rsFC alterations within the advanced FNM, especially the increased rsFC within thalamus-insula loop, suggesting that excessive sensitivity to external information plays an important role in PD. The advanced FNM may provide a fuller explanation about PD.

SFC-HO: Reliable Layered Service Function Chaining

Network function virtualization (NFV) utilizes IT virtualization technology to realize the functions of various network devices and realize the decoupling of hardware and software functions. Typically, a service request specifies the virtual network functions (VNFs) it needs and the order between them. A network flow needs to traverse a set of sequential VNFs called a service function chain (SFC). Although SFC can increase the flexibility of service orchestration to meet the needs of different users, network providers face many challenges due to the need to ensure service reliability and some constraints. Therefore, how orchestrating SFC and designing a suitable network architecture is a critical issue in this field at present. We propose an efficient network layer-based SFC orchestration method, called SFC-hierarchical orchestration (SFC-HO). Our method separates virtual and physical networks into layers and computes the availability of resources in each layer. We filter the VNFs of each layer and deploy them to the physical layer that meets the greatest benefit. To this end, we formulate the problem as an integer linear programming (ILP) problem to minimize the total deployment cost. In order to further optimize the layering strategy, we innovatively use the Benders decomposition method to decompose the SFC-HO problem into two sub-problems, which are the hierarchical mapping problem of VNFs and the routing problem between nodes. We propose two algorithms for the two problems respectively. The simulation results show that compared with the SFC orchestration process in the traditional network model. Our research can effectively improve the reliability of the SFC, reduce the delay of the routing process, and effectively reduce the cost consumption. Our research effectively solves the problem of difficult service orchestration for operators in the face of diverse service demands and a large number of service requests.

Thermoacoustic stability prediction using classification algorithms

Abstract Predicting the occurrence of thermoacoustic instabilities is of major interest in a variety of engineering applications such as aircraft propulsion, power generation, and industrial heating. Predictive methodologies based on a physical approach have been developed in the past decades, but have a moderate-to-high computational cost when exploring a large number of designs. In this study, the stability prediction capabilities and computational cost of four well-established classification algorithms—the K-Nearest Neighbors, Decision Tree (DT), Random Forest (RF), and Multilayer Perceptron (MLP) algorithms—are investigated. These algorithms are trained using an in-house physics-based low-order network model tool called OSCILOS. All four algorithms are able to predict which configurations are thermoacoustically unstable with a very high accuracy and a very low runtime. Furthermore, the frequency intervals containing unstable modes for a given configuration are also accurately predicted using multilabel classification. The RF algorithm correctly predicts the overall stability and finds all frequency intervals containing unstable modes for 99.6 and 98.3% of all configurations, respectively, which makes it the most accurate algorithm when a large number of training examples is available. For smaller training sets, the MLP algorithm becomes the most accurate algorithm. The DT algorithm is found to be slightly less accurate, but can be trained extremely quickly and runs about a million times faster than a traditional physics-based low-order network model tool. These findings could be used to devise a new generation of combustor optimization tools that would run much faster than existing codes while retaining a similar accuracy.

Community Detection in Fully-Connected Multi-layer Networks Through Joint Nonnegative Matrix Factorization

Modern data analysis and processing tasks typically involve large sets of structured data. Graphs provide a powerful tool to describe the structure of such data, where the entities and the relationships between them are modeled as the nodes and edges of the graph. Traditional single layer network models are insufficient for describing the multiple entity types and modes of interaction encountered in real-world applications. Recently, multi-layer network models, which consider the different types of interactions both within and across layers, have emerged to model these systems. One of the important tools in understanding the topology of these high-dimensional networks is community detection. In this paper, a joint nonnegative matrix factorization approach is proposed to detect the community structure in multi-layer networks. The proposed approach models the multi-layer network as the union of a multiplex and bipartite network and formulates community detection as a regularized optimization problem. This optimization problem simultaneously finds the nonnegative low-rank embedding of the intra- and inter-layer adjacency matrices while minimizing the distance between the two to guarantee pair-wise similarity across embeddings. The proposed approach can detect the community structure for both homogeneous and heterogeneous multi-layer networks and is robust to noise and sparsity. The performance of the proposed approach is evaluated for both simulated and real networks and compared to state-of-the-art methods.

Product Competition Prediction in Engineering Design Using Graph Neural Networks

AbstractUnderstanding relationships between different products in a market system and predicting how changes in design impact their market position can be instrumental for companies to create better products. We propose a graph neural network-based method for modeling relationships between products, where nodes in a network represent products and edges represent their relationships. Our modeling enables a systematic way to predict the relationship links between unseen products for future years. When applied to a Chinese car market case study, our method based on an inductive graph neural network approach, GraphSAGE, yields double the link prediction performance compared to an existing network modeling method—exponential random graph model-based method for predicting the car co-consideration relationships. Our work also overcomes scalability and multiple data type-related limitations of the traditional network modeling methods by modeling a larger number of attributes, mixed categorical and numerical attributes, and unseen products. While a vanilla GraphSAGE requires a partial network to make predictions, we augment it with an “adjacency prediction model” to circumvent the limitation of needing neighborhood information. Finally, we demonstrate how insights obtained from a permutation-based interpretability analysis can help a manufacturer understand how design attributes impact the predictions of product relationships. Overall, this work provides a systematic data-driven method to predict the relationships between products in a complex network such as the car market.

DR-IIXRN : Detection Algorithm of Diabetic Retinopathy Based on Deep Ensemble Learning and Attention Mechanism.

Diabetic retinopathy (DR) is one of the common chronic complications of diabetes and the most common blinding eye disease. If not treated in time, it might lead to visual impairment and even blindness in severe cases. Therefore, this article proposes an algorithm for detecting diabetic retinopathy based on deep ensemble learning and attention mechanism. First, image samples were preprocessed and enhanced to obtain high quality image data. Second, in order to improve the adaptability and accuracy of the detection algorithm, we constructed a holistic detection model DR-IIXRN, which consists of Inception V3, InceptionResNet V2, Xception, ResNeXt101, and NASNetLarge. For each base classifier, we modified the network model using transfer learning, fine-tuning, and attention mechanisms to improve its ability to detect DR. Finally, a weighted voting algorithm was used to determine which category (normal, mild, moderate, severe, or proliferative DR) the images belonged to. We also tuned the trained network model on the hospital data, and the real test samples in the hospital also confirmed the advantages of the algorithm in the detection of the diabetic retina. Experiments show that compared with the traditional single network model detection algorithm, the auc, accuracy, and recall rate of the proposed method are improved to 95, 92, and 92%, respectively, which proves the adaptability and correctness of the proposed method.

Non-Isothermal Optimal Power and Gas Flow

Isothermal natural gas flow model with constant gas temperature has been widely used in conventional coordinated operations of integrated electricity and natural gas systems (IEGS). However, as a crucial state variable in IEGS, gas temperature varies significantly among nodes in a natural gas system, which can be calculated along with nodal pressures and mass flow rates. This paper proposes a novel non-isothermal optimal power and gas flow (OPGF) model that avoids the isothermal assumptions involved in traditional natural gas network model and improves the accuracy significantly. In the proposed model, partial differential terms with minimal weights are neglected based on the quantitative analysis of the energy conservation equation to further simplify the energy conservation equation. A non-isothermal model of natural gas networks is developed by introducing the finite difference method and laws of thermodynamics to incorporate the gas temperature change into the coordinated operation of IEGS. A start point initialization scheme is developed based on the Newton-Raphson algorithm to accelerate the interior point method solution convergence and improve its solution quality. Numerical experiments presented for two test systems validate the effectiveness of the proposed OPGF model and solution methodology.

An Apple Detection Method Based on Des-YOLO v4 Algorithm for Harvesting Robots in Complex Environment

Real-time detection of apples in natural environment is a necessary condition for robots to pick apples automatically, and it is also a key technique for orchard yield prediction and fine management. To make the harvesting robots detect apples quickly and accurately in complex environment, a Des-YOLO v4 algorithm and a detection method of apples are proposed. Compared with the current mainstream detection algorithms, YOLO v4 has better detection performance. However, the complex network structure of YOLO v4 will reduce the picking efficiency of the robot. Therefore, a Des-YOLO structure is proposed, which reduces network parameters and improves the detection speed of the algorithm. In the training phase, the imbalance of positive and negative samples will cause false detection of apples. To solve the above problem, a class loss function based on AP-Loss (Average Precision Loss) is proposed to improve the accuracy of apple recognition. Traditional YOLO algorithm uses NMS (Nonmaximum Suppression) method to filter the prediction boxes, but NMS cannot detect the adjacent apples when they overlap each other. Therefore, Soft-NMS is used instead of NMS to solve the problem of missing detection, so as to improve the generalization of the algorithm. The proposed algorithm is tested on the self-made apple image data set. The results show that Des-YOLO v4 network has ideal features with a mAP (mean Average Precision) of apple detection of 97.13%, a recall rate of 90%, and a detection speed of 51 f/s. Compared with traditional network models such as YOLO v4 and Faster R-CNN, the Des-YOLO v4 can meet the accuracy and speed requirements of apple detection at the same time. Finally, the self-designed apple-harvesting robot is used to carry out the harvesting experiment. The experiment shows that the harvesting time is 8.7 seconds and the successful harvesting rate of the robot is 92.9%. Therefore, the proposed apple detection method has the advantages of higher recognition accuracy and faster recognition speed. It can provide new solutions for apple-harvesting robots and new ideas for smart agriculture.

Face mask detection algorithm for multi task recognition in complex environment

The right wearing masks can effectively prevent the spread of viruses. There are usually complex distractions in densely populated environments, which will affect detection tasks of face masks. However, the current detecting algorithm based on Faster R-CNN can not meet mask wearing detection of small targets in complex environments, so this paper puts forward a mask wearing detection algorithm based on the improved Faster R-CNN. This algorithm changes the traditional single RPN network model into one multi-task enhanced RPN model, which can improve the detection and recognition accuracy, the improved Soft-NMS algorithm is used to delete the redundant candidate boxes of regional candidate network output. At the same time, CIoU loss function with better performance, which fully considers the center point distance between the target and the detection frame and improves the accuracy of detection, replaced the original IoU loss function. 3000 sample images are used to do training and experimental verification of the model according to 1:3ratio of positive and negative samples. Experimental results show that compared with the Faster R-CNN algorithm, the algorithm proposed in this paper improves the face detection accuracy and face mask wearing detection by 7.15% and 15.99%respectively. Besides, the average detection accuracy improved by 11.57% and the FPS by 4.6.

GPU accelerated power flow calculation of integrated electricity and heat system with component-oriented modeling of district heating network

Due to its advantage of improving energy efficiency and promoting sustainable development, integrated electricity and heat system (IEHS) has been widely studied in recent decades. However, the traditional network-oriented district heating network (DHN) model in IEHS could only deal with DHNs of supply-return-parallel topologies, and the employ of constant thermodynamic properties could incur inaccurate power flow results. With the increasing requirements on operation flexibility and system resilience of IEHS, it has become a necessity to develop a superior power flow model of DHN. This study presents a novel component-oriented modeling method in which the models of the three basic components in DHN, the pipelines, pressure sources and junctions, are investigated in detail. Formulas of the fundamental physical processes including pressure, temperature loss and enthalpy transfer are derived based on the variable thermodynamic state of the fluid rather than predetermined constants in the traditional simplified models. Variables associated with these basic components are discussed in detail and their respective constraints are expounded. To overcome the huge amount of computation in the IEHS analyzing process, GPU is introduced as a coprocessor and a parallel algorithm is designed accordingly. The versatility of the proposed model, including providing accurate, more detailed power flow results and analyzing DHN of general topologies, is presented in a small-scale DHN case. And the practicality of the proposed model is demonstrated in the ensuing practical-scale IEHS case. Meanwhile, the proposed GPU-based parallel algorithm has attained more than 3 times of performance boost compared to single CPU computing.

Spoofing Attack Detection Using Machine Learning in Cross-Technology Communication

Cross-technology communication (CTC) technique can realize direct communication among heterogeneous wireless devices (e.g., WiFi, ZigBee, and Bluetooth in the 2.4 G ISM band) without gateway equipment for forwarding, which makes heterogeneous wireless communication more convenient and greatly reduces communication costs. However, compared with the traditional homogeneous network model, CTC technique also makes it easier to implement spoofing attacks in heterogeneous networks. WiFi devices with long communication distances and sufficient energy supply can directly launch spoofing attacks against ZigBee devices, which brings severe security concerns for heterogeneous wireless communications. In this paper, we focus on the CTC spoofing attack, especially spoofing attacks from WiFi to ZigBee and propose a machine learning-based method to detect spoofing attacks for heterogeneous wireless networks by using physical-layer information. First, we model the received signal strength (RSS) data of legitimate ZigBee devices to construct a one-class support vector machine (OSVM) classifier for detecting CTC spoofing attacks depending on the obtained training samples. Then, we simulated CTC spoofing attacks in a live testbed and evaluated the performance of our detection method. Results show that our approach is highly effective in spoofing detection. Even if the distance between the legitimate ZigBee device and WiFi attacker is near each other (i.e., less than 2 m) and does not require a large number of samples, the detection rate and precision of our method are both over 90%. Finally, we employ the OSVM classifier to obtain samples of spoofing attacks and then explore using SVM to further improve the performance of the classifier.

Online preschool education optimization based on edge computing in the era of COVID‐19

Under the influence of COVID‐19, people's normal life and activities have been limited, such as the education of children, which leads to the emergence of online preschool education. Since online preschool education is large‐scale and time‐sensitive, the traditional network model cannot satisfy the needs of online education. In this paper, edge computing is adopted to optimize online preschool education, where a task unloading algorithm based on genetic algorithm (TUOGA) is designed to minimize the computing delay of terminal tasks. In order to verify the effectiveness and efficiency of the proposed algorithm, TUOGA is compared with two task offloading algorithms, and simulation results show that the proposed algorithm outperforms them in the aspect of time latency.

Rapid Modeling of the Sound Speed Field in the South China Sea Based on a Comprehensive Optimal LM-BP Artificial Neural Network

Ocean sound speed is an essential foundation for marine scientific research and marine engineering applications. In this article, a model based on a comprehensive optimal back propagation artificial neural network model is developed. The Levenberg–Marquardt algorithm is used to optimize the model, and the momentum term, normalization, and early termination method were used to predict the high precision marine sound speed profile. The sound speed profile was described by five indicators: date, time, latitude, longitude, and depth. The model used data from the CTD observation dataset of scientific investigation over the South China Sea (2009–2012) (108°–120° E, 6°–8° N), which includes comprehensive scientific investigation data from four voyages. The feasibility of modeling the sound speed field in the South China Sea is investigated. The proposed model uses the momentum term, normalization, and early termination in a traditional BP artificial neural network structure and mitigates issues with overtraining and difficulty when determining the BP neural network parameters. With the LM algorithm, a fast-modeling method for the sound field effectively achieves the precision requirement for sound speed prediction. Through the prediction and verification of the data from 2009 to 2012, the newly proposed optimized BP network model is shown to dramatically reduce the training time and improve precision compared to the traditional network model. Results showed that the root mean squared error decreased from 1.7903 m/s to 0.95732 m/s, and the training time decreased from 612.43 s to 4.231 s. Finally, the sound ray tracing simulations confirm that the model meets the accuracy requirements of acoustic sounding and verify the model’s feasibility for the real-time prediction of the vertical sound speed in saltwater bodies.

Application of quantum-like Bayesian network and belief entropy for interference effect in multi-attribute decision making problem

Multi-attribute decision making (MADM) is a classical model that widely used in various fields. But most research about this topic often assume that the attributes are independent each other. In fact the attributes maybe interact or interference each other in many cases. There are researches on the interact relationships with Choquet integral, but there is no research on the interference effect by far to the authors. In this paper, a multi-attribute decision-making model is proposed by considering the interference effects between attributes with quantum-like Bayesian network (QLBN) and belief entropy. Quantum probability theory (QPT) is more in line with people's thinking by considering the interference relationships between attributes, which is represented as the interference value in quantum theory. Belief entropy, an index to calculate the uncertainty of probability, is introduced as a method to calculate the interference value in this paper. Finally, the model is applied to assessing the industrial economic benefit as well as the quantitative performance comparison between the traditional probability and the QLBN probability is presented. It shows that the combination of QLBN model and belief entropy has prediction ability compared with the traditional Bayesian network model.

Sonar Image Target Detection and Recognition Based on Convolution Neural Network

Recent advancements in deep learning offer an effective approach for the study in machine vision using optical images. In this paper, a convolution neural network is used to deal with the target task of sonar detection, and the performance of each neural network model in the sonar image detection and recognition task of underwater box and tire is compared. The simulation results show that the neural network method proposed in this paper is better than the traditional machine learning methods and SSD network models. The average accuracy of the proposed method for sonar image target recognition is 93%, and the detection time of a single image is only 0.3 seconds.

Trends and Recent Change in the Syntactic Distribution of Degree Modifiers: Implications for a Usage-based Theory of Word Classes

This paper examines the syntactic distribution of degree modifiers in both spoken and written English. The results of the empirical case studies show that degree modifiers, both amplifiers (e.g., very, extremely) and downtoners (e.g., quite, pretty), are generally more often used in predication than in attribution, a result that is in line with earlier observations of the distribution of individual modifiers. This synchronic trend is also evident in diachronic developments: corpus data show that the recent frequency increase of intensifying this and that has largely taken place in predication, and the adjectivization of a class of -ed participles (e.g., interested, scared) can also be connected to their frequent co-occurrence with degree modifiers after be. Finally, the connection between degree modifiers and predicative usage has recently become stronger for a subset of modifiers (e.g., so, this, that) due to the decline of the “Big Mess” construction (e.g., so good an idea). From a theoretical perspective, this paper promotes a dynamic, usage-based model of word classes where frequency of use plays a role in categorization. The data investigated in the article are mainly discussed from the perspective of usage-based Construction Grammar, and the theoretical implications of the findings are examined both in light of a more traditional Construction Grammar network model of language and some recent ideas of overlapping word classes.

The aging effect in evolving scientific citation networks

The study of citation networks is of interest to the scientific community. However, the underlying mechanism driving individual citation behavior remains imperfectly understood, despite the recent proliferation of quantitative research methods. Traditional network models normally use graph theory to consider articles as nodes and citations as pairwise relationships between them. In this paper, we propose an alternative evolutionary model based on hypergraph theory in which one hyperedge can have an arbitrary number of nodes, combined with an aging effect to reflect the temporal dynamics of scientific citation behavior. Both theoretical approximate solution and simulation analysis of the model are developed and validated using two benchmark datasets from different disciplines, i.e. publications of the American Physical Society (APS) and the Digital Bibliography & Library Project (DBLP). Further analysis indicates that the attraction of early publications will decay exponentially. Moreover, the experimental results show that the aging effect indeed has a significant influence on the description of collective citation patterns. Shedding light on the complex dynamics driving these mechanisms facilitates the understanding of the laws governing scientific evolution and the quantitative evaluation of scientific outputs.

Formation of Organizational and Economic Model of Cross-Industry Ecosystems

The study focuses on the research of fundamental reasons underlying the emergence of digital platform providing the formation of the global view on the role of digital platforms for building of a new organizational structure - “high-tech industrial ecosystem”.The purpose of the study is the validation of development a modern mechanism for the coordination of high-tech market entities within a single economic and organizational space – the ecosystem based upon the cross-industry digital platform.This study, using general methods of scientific knowledge in various aspects, analyses the modern vector of the industrial growth explained by the implementation of ecosystems as new organizational and economic models. The principles for their formation, the probable structure, main differences from the traditional cluster and network models are reflected. It is proved that it is the ecosystem model that makes the ecosystem members achieve a positive synergetic effect during the implementation of their strategic growth targets in the conditions of digital transformation.The article considers the application of tools of the industrial digital platform providing for the implementation of interaction between ecosystem members. Platform-based solutions have the big future from the viewpoint of the large data arrays analysis, reduced transaction losses and “perfect information” obtaining. As a result, the immediate course for introduction of cross-industry digital platform and ecosystem building is justified.Digital cross-industry interaction within the platform will help to expand external communication and promotion channels, introduce digital business models and diversify production but it requires the provision of compatibility between the systems of industrial companies, functioning of a single digital platform and the cloud-based environment.

Flower image classification based on generative adversarial network and transfer learning

Aiming at the problem that the classification accuracy of the traditional flower classification method is low and the deep neural network requires a large amount of original data. This paper designs a flower classification model that combines generative adversarial network and ResNet-101 transfer learning algorithm, and uses stochastic gradient descent algorithm to optimize the training process of the model. The experimental results on the the international public flower recognition dataset, Oxford flower-102 dataset, show that by enhancing the original data, the accuracy of the network's recognition and classification of flowers is improved. At the same time, the model proposed in this paper is superior to other traditional network models, with higher recognition accuracy and robustness.

Vehicle detection algorithm based on LW-SSD

Due to the complex structure of the traditional SSD network model and the large amount of parameters, it is not suitable for porting to mobile or embedded platforms, so the LW-SSD vehicle detection algorithm is proposed, this paper redesigns the basic feature extraction network of SSD, using MobileNet feature extraction network instead of VGG16, for the original MobileNet to improve, remove the original MobileNet network average pooling layer and Softmax layer, while reducing the MobileNet network from the original 28 layers to 13 layers. Then, on the basis of the subsequent network, feature maps of different scales are scaled, residual mapping and deep feature fusion operations, and the low-level position information and high-level semantic information are effectively combined to obtain new feature maps of different scales. The designed network model is trained using the Tensorflow framework, and the test effect of the network model is evaluated. The experimental results show that the accuracy of the network model designed in this paper is 78.76%, which is 9.27% lower than the original SSD network model. However, the amount of parameters is only 1/4 of the SSD network model. At the expense of partial accuracy, the design of a lightweight network model is exchanged, which lays the foundation for subsequent development on mobile or embedded platforms.