Network Structure Data Research Articles

A-star algorithm vs. Q-learning algorithm for path planning: A comparative analysis

In robotics research, the core focus is on path planning, as it's essential for robots to navigate according to human objectives to be valuable. This paper provides an in-depth analysis of the commonly used A-Star algorithm and Q-learning algorithm in mobile robot path planning. It specifically examines the advantages and disadvantages of both the A-Star algorithm and the Q-learning algorithm, highlighting their roles in the advancement of robotic intelligence. The A-star algorithm is an algorithm with a heuristic function to guide path selection direction. Also, the A-star algorithm is guaranteed to identify the most efficient path in terms of traffic cost when facing static obstacles. However, its performance degrades when dealing with trap nodes. Specifically, the algorithm occupies a large amount of memory and takes a long computational time. In contrast, the Q-learning algorithm is effective for learning continuous actions with the same network data structure for local path planning tasks.

The Use of Generative Adversarial Network and Graph Convolution Network for Neuroimaging-Based Diagnostic Classification.

Functional connectivity (FC) obtained from resting-state functional magnetic resonance imaging has been integrated with machine learning algorithms to deliver consistent and reliable brain disease classification outcomes. However, in classical learning procedures, custom-built specialized feature selection techniques are typically used to filter out uninformative features from FC patterns to generalize efficiently on the datasets. The ability of convolutional neural networks (CNN) and other deep learning models to extract informative features from data with grid structure (such as images) has led to the surge in popularity of these techniques. However, the designs of many existing CNN models still fail to exploit the relationships between entities of graph-structure data (such as networks). Therefore, graph convolution network (GCN) has been suggested as a means for uncovering the intricate structure of brain network data, which has the potential to substantially improve classification accuracy. Furthermore, overfitting in classifiers can be largely attributed to the limited number of available training samples. Recently, the generative adversarial network (GAN) has been widely used in the medical field for its generative aspect that can generate synthesis images to cope with the problems of data scarcity and patient privacy. In our previous work, GCN and GAN have been designed to investigate FC patterns to perform diagnosis tasks, and their effectiveness has been tested on the ABIDE-I dataset. In this paper, the models will be further applied to FC data derived from more public datasets (ADHD, ABIDE-II, and ADNI) and our in-house dataset (PTSD) to justify their generalization on all types of data. The results of a number of experiments show the powerful characteristic of GAN to mimic FC data to achieve high performance in disease prediction. When employing GAN for data augmentation, the diagnostic accuracy across ADHD-200, ABIDE-II, and ADNI datasets surpasses that of other machine learning models, including results achieved with BrainNetCNN. Specifically, in ADHD, the accuracy increased from 67.74% to 73.96% with GAN, in ABIDE-II from 70.36% to 77.40%, and in ADNI, reaching 52.84% and 88.56% for multiclass and binary classification, respectively. GCN also obtains decent results, with the best accuracy in ADHD datasets at 71.38% for multinomial and 75% for binary classification, respectively, and the second-best accuracy in the ABIDE-II dataset (72.28% and 75.16%, respectively). Both GAN and GCN achieved the highest accuracy for the PTSD dataset, reaching 97.76%. However, there are still some limitations that can be improved. Both methods have many opportunities for the prediction and diagnosis of diseases.

Enhancing Threat Prioritization In Cybersecurity Using Avian Shield Optimizer (ASO) In Graph Forge Elite (GFE) Framework

In the ever-evolving field of cybersecurity, the ability to effectively prioritize threats is essential for organizations to allocate resources wisely and proactively mitigate potential risks. Avian Shield Optimizer (ASO) is a revolutionary optimization technique specifically designed for threat prioritization. ASO utilizes evolutionary-inspired mechanisms to dynamically adjust the prioritization of threats based on a comprehensive analysis of factors such as severity, relevance, and potential impact on organizational assets. By continuously monitoring and adapting to the dynamic threat landscape, ASO enables organizations to anticipate and respond quickly to emerging threats, ensuring that resources are allocated appropriately to address the most critical vulnerabilities. Moreover, ASO seamlessly integrates with existing threat intelligence platforms, providing cybersecurity teams with actionable insights and recommendations to comprehensively strengthen their defense strategies. Concurrently, the GraphForgeElite framework enhances this approach by facilitating the analysis of complex network data connections and structures, enabling the detection of subtle irregularities that may indicate potential cyber threats. ASO collaborates synergistically with GraphForgeElite, utilizing evolutionary-inspired methodologies to dynamically adapt the framework's design and settings, automatically tuning the parameters. Through rigorous experimentation, the performance of ASO-GraphForgeElite's Network is compared to other state-of-the-art classifiers. The results demonstrate the superior performance of ASO-GraphForgeElite's Network, surpassing 99% accuracy, precision, recall, and F1-score. Additionally, the framework exhibits efficiency in handling complex network data structures, enabling the identification of subtle patterns that indicate potential threats.

Целостность и иерархичность структур данных в жизненном цикле объектов малоэтажного жилищного строительства

The subject area of the article is construction facilities life cycle management, where the object-oriented automated technologies of low-rise housing construction are the subject of consideration. The purpose of this study is to describe the meta-structure of such technologies, consisting of a set of hierarchical structures of the finite elements (premises, structures, engineering equipment, etc.) and for this meta-structure definition the concept of low-rise housing estate construction facilities integrity is introduced. The object of the article's research is hierarchical data structures, on the basis of which object-oriented databases and knowledge bases are formed, serving automated technologies for construction facilities integrity generation within all stages of the life cycle. The article uses the method of decomposition of the low-rise housing estate construction facilities integrity into a variety of automated technologies and their constituent hierarchical data structures, which allows us to identify all subject areas of objects of manor low-rise housing estate construction for computational justifications of the limiting states of finite elements (objects) of architecture, building structures, engineering equipment. Hierarchical structures create a standardized basis for the organization of finite element information models for managing the life cycle of manor low-rise estate housing construction. Decision-makers should prepare hierarchical data structures and algorithms for calculating the justifications for finite elements. The article presents examples of hierarchical data structures schemes for the design stage tasks solving. It is concluded that all information models of finite elements on a variety of subject areas of manor low-rise estate housing construction objects arise during the execution of constructive (algorithmized) processes in the decision-making layers of the corresponding stages of the life cycle of construction objects. It is concluded that all information models of finite elements on a variety of subject areas of manor low-rise estate housing construction objects arise during the execution of constructive (algorithmized) processes in the decision-making layers of the corresponding stages of the life cycle of construction objects. A structuring condition is also proposed: if the main link of the life cycle management technology is the information model of the object, then the fundamental condition for object orientation and meaningful filling of hierarchical data structures is the construction system used in this construction object, as a set of load-bearing and enclosing structures, its components and materials used. The proposed approach will make it possible to create and maintain a shared data environment on a regular and ongoing basis as part of a set of hierarchical data structures of the construction projects under consideration. In the development, it becomes possible to qualitatively improve the technological conditions for decision-making according to various criteria, including criteria for environmental friendliness, energy efficiency, economy, comfort for low-rise residential buildings using neural network data structures and the use of artificial intelligence algorithms. The above together is the basis for the formation of a modern innovative technological order for the objects of estate low-rise housing construction.

NodeVector: A Novel Network Node Vectorization with Graph Analysis and Deep Learning

Network node embedding captures structural and relational information of nodes in the network and allows for us to use machine learning algorithms for various prediction tasks on network data that have an inherently complex and disordered structure. Network node embedding should preserve as much information as possible about important network properties where information is stored, such as network structure and node properties, while representing nodes as numerical vectors in a lower-dimensional space than the original higher dimensional space. Superior node embedding algorithms are a powerful tool for machine learning with effective and efficient node representation. Recent research in representation learning has led to significant advances in automating features through unsupervised learning, inspired by advances in natural language processing. Here, we seek to improve the representation quality of node embeddings with a new node vectorization technique that uses network analysis to overcome network-based information loss. In this study, we introduce the NodeVector algorithm, which combines network analysis and neural networks to transfer information from the target network to node embedding. As a proof of concept, our experiments performed on different categories of network datasets showed that our method achieves better results than its competitors for target networks. This is the first study to produce node representation by unsupervised learning using the combination of network analysis and neural networks to consider network data structure. Based on experimental results, the use of network analysis, complex initial node representation, balanced negative sampling, and neural networks has a positive effect on the representation quality of network node embedding.

Multistage reliability-constrained stochastic planning of diamond distribution network: An approximate dynamic programming approach

In this paper, a multistage reliability-constrained stochastic planning model is established for the diamond distribution network, a new type of distribution system grid structure, with the goal of optimizing the total cost of investment and operation. The uncertainty of load growth is simulated by generating multiple load growth scenarios, and the power supply loss caused by the reliability of components in the system is also considered. The multistage stochastic dynamic decision process is modeled by The Markov decision process. In order to solve the “curse of dimensionality” problem in multi-stage stochastic planning model, the approximate dynamic programming algorithm is used to solve it. The case study is based on the typical structure of the diamond distribution network and realistic data of an actual district in Shanghai, China. The planning results verify the effectiveness of the model and the solution algorithm, and demonstrate the reliability and economy of the diamond distribution network are superior to the traditional double-ring network.

An analytic strategy for data processing of multimode networks

Complex network data structures are considered to capture the richness of social phenomena and real-life data settings. Multipartite networks are an example in which various scenarios are represented by different types of relations, actors, or modes. Within this context, the present contribution aims at discussing an analytic strategy for simplifying multipartite networks in which different sets of nodes are linked. By considering the connection of multimode networks and hypergraphs as theoretical concepts, a three-step procedure is introduced to simplify, normalize, and filter network data structures. Thus, a model-based approach is introduced for derived bipartite weighted networks in order to extract statistically significant links. The usefulness of the strategy is demonstrated in handling two application fields, that is, intranational student mobility in higher education and research collaboration in European framework programs. Finally, both examples are explored using community detection algorithms to determine the presence of groups by mixing up different modes.

Community Detection in Multilayer Networks Based on Matrix Factorization and Spectral Embedding Method

Community detection remains a challenging research hotspot in network analysis. With the complexity of the network data structures increasing, multilayer networks, in which entities interact through multiple types of connections, prove to be effective in describing complex networks. The layers in a multilayer network may not share a common community structure. In this paper, we propose a joint method based on matrix factorization and spectral embedding to recover the groups not only for the layers but also for nodes. Specifically, the layers are grouped via the matrix factorization method with layer similarity-based regularization in the perspective of a mixture multilayer stochastic block model, and then the node communities within a layer group are revealed by clustering a combination of the spectral embedding derived from the adjacency matrices and the shared approximation matrix. Numerical studies show that the proposed method achieves competitive clustering results as the number of nodes and/or number of layers vary, together with different topologies of network layers. Additionally, we apply the proposed method on two real-world multilayer networks and obtain interesting findings which again highlight the effectiveness of our method.

Design and Development of a Data Structure Visualisation System Using the Ant Colony Algorithm

Aim: A data structure visualisation system uses object-oriented thinking and COM technology to dynamically and interactively simulate and track data structure algorithms and realize the dynamic synchronisation and visualisation of abstract data structures and algorithms using the data modelling function and self-test function. Background: Teaching data structures and algorithms is difficult because of their abstraction and dynamics; the use of icons in classroom teaching can be partially abstracted into intuition, but analysing the instantaneous dynamic characteristics of the object and the dynamic execution process of the algorithm is difficult. Objective: A data structure visualisation system employs a data modelling function, visual data structures, a user-friendly and flexible interface, and multipath features for multilevel users. Such systems can be designed more effectively by using the ant colony algorithm. Methods: The more the ants pass by a certain path, the higher the concentration of residual pheromone, and the higher the probability of the subsequent ant selecting that path. Therefore, the individuals in an ant colony communicate messages and cooperate with each other for foraging. Results: The resulting speedup ratio indicates that the speedup is smaller when the number of nodes is 100 or more; the acceleration is higher when the node reaches a certain scale, and the speedup ratio does not change considerably. In this study, the traffic simulation software VISSIM was used to generate road network data; the generated traffic data were analysed and used to design a traffic network data structure. Conclusion: The traffic network data model oriented to the ant colony algorithm was established through abstraction. Accordingly, a parallel ant colony algorithm based on cloud computing was implemented. Finally, a Hadoop cloud computing platform was established and used to run and test the parallel ant colony algorithm program; several experiments were conducted, and the experimental results were analysed concurrently.

DDPLA: A Dynamic Differential Privacy Algorithm for Social Network Based on Local Community

<p>Social networks contain a large number of privacy information. Personal privacy will be jeopardized if network data without privacy protection is released directly. In view of the current privacy protection technology to protect the social network, there are some problems such as low maintenance of network structure or low accuracy of network data. In order to solve these problems, this paper proposes a dynamic differential privacy algorithm for social network based on local community (DDPLA). The algorithm can divide the social network into different communities, dynamically generate privacy budgets for different communities, and then generate uncertainty graphs. Experiments show that compared with other algorithms, the social network processed by DDPLA algorithm can better balance data utility and privacy protection. Furthermore, the algorithm can better protect important nodes.</p> <p> </p>

Three-Dimensional Tetrahedron Network (3D TEN) and Volumetric Geo-Objects (VGO) Data Model for Groundwater Simulation in Bida Basin, Nigeria

The presence of complex ground soil structures limits the attempt for groundwater extraction for alternate water supply usage. Hydrologists and engineers have made many attempts to Model 3D groundwater flow and storage using digital elevation model (DEM) data structures but have resulted in limited success in terms of actual 3D subsurface geology representation. This study attempts to construct resistivity data in an actual 3D environment using volumetric geo-objects (VGOs) and 3D tetrahedron network (3D TEN) data structures. Subsurface data were collected using the vertical electrical sounding (VES) approach to analyse the geological structure of the study area. To determine the course and movement of groundwater, the 3D constant was applied from Darcy's law in the Lagrangian technique. A time frame of one hour, two hours, four hours and eight-hour intervals was tested to simulate the porosity and permeability of various soil types with complex geological structures. The findings validated the use of 3D visualisation to model resistivity data. The implemented volumetric geo-object (VGO) data model and the 3D TEN model show the volume of water over the topsoil to be 17,461,228.07 in an hour; clay soil was 870,614.04; clayey sand was 435,307.02; and sandy clays was 348,245.61 per kilometre cube (km3). The use of 3D TEN and VGO for groundwater simulation in this research offers an improved perspective in groundwater visualisation that will promote sustainable groundwater exploration for societal developments. The findings of this study will be beneficial to stakeholders, water resource managers and government agencies such as Ministry of Water Resources. Keywords: Data model, Groundwater simulation, Three-dimensional tetrahedron network (3D TEN), Vertical electrical sounding (VES), Visualisation, Volumetric geo-objects (VGOs)

Constraining the Charm-Yukawa coupling at the Large Hadron Collider

We present theoretical results for the sensitivity of charm Yukawa coupling measurements in future high-luminosity LHC runs in three channels: Vector Boson Fusion (VBF), W Higgs-strahlung and Z Higgs-strahlung production of a Higgs boson and its subsequent decay into charm quarks. To reduce the overwhelmingly large backgrounds and to reduce false positives, we apply a set of simple kinematic and jet feature cuts and feed neural network data structures of three types; jet features, jet images and particle level features. To facilitate straightforward comparison with experimental studies [1,2], we express our results in terms of signal strengths [3].

Statistical Network Analysis with Bergm

Recent advances in computational methods for intractable models have made network data increasingly amenable to statistical analysis. Exponential random graph models (ERGMs) emerged as one of the main families of models capable of capturing the complex dependence structure of network data in a wide range of applied contexts. The Bergm package for R has become a popular package to carry out Bayesian parameter inference, missing data imputation, model selection and goodness-of-fit diagnostics for ERGMs. Over the last few years, the package has been considerably improved in terms of efficiency by adopting some of the state-of-the-art Bayesian computational methods for doublyintractable distributions. Recently, version 5 of the package has been made available on CRAN having undergone a substantial makeover, which has made it more accessible and easy to use for practitioners. New functions include data augmentation procedures based on the approximate exchange algorithm for dealing with missing data, adjusted pseudolikelihood and pseudo-posterior procedures, which allow for fast approximate inference of the ERGM parameter posterior and model evidence for networks on several thousands nodes.

Systematic mapping of R-group space enables the generation of an R-group replacement system for medicinal chemistry.

Selection of R-groups (substituents, functional groups) is of critical importance for the generation of analogues during hit-to-lead and lead optimization. In the practice of medicinal chemistry, R-group selection is mostly driven by chemical experience and intuition taking synthetic criteria into account. However, systematic analyses of substituents are currently rare. In this work, we have computationally isolated R-groups from more than 17,000 analog series comprising ∼315,000 bioactive compounds. From more than 50,000 unique substituents, frequently used R-groups were identified. For these R-groups, preferred replacements over more than 60,000 individual substitution sites were identified with the aid of a network data structure. These data provided the basis for the generation of a searchable R-group replacement system for medicinal chemistry containing replacement hierarchies for frequently used R-groups, which is made freely available as the central component of our study.

Type-2 Fuzzy Neural System for Diagnosis of Diabetes

Diabetes is a chronic disease that is characterized by insufficient production or utilization of insulin and a consequent high increase in blood sugar. Diagnosis of diabetes is a complex process and requires a high level of expertise. The disease is characterized by a set of signs and symptoms. Some of these symptoms are obtained through laboratory analysis. Creation of a knowledge base and automation of disease diagnosis are important and allow fast detection and treatment. Various techniques have been used to develop a high-accuracy system for the diagnosis of diabetes. Fuzzy logic is one of the appropriate methodologies for the development of such medical diagnostic systems. Several research studies have used fuzzy models to diagnose medical diseases due to the imprecision and uncertainty associated with medical data. Moreover, a high level of uncertainty in medical data requires a type-2 fuzzy system to handle these uncertainties and diagnose diabetes. The paper proposes the integration of a type-2 fuzzy system and neural networks for the diagnosis of diabetes. Using the structure of type-2 fuzzy neural network (T2FNN) and statistical data, the system’s design for the diagnosis of diabetes is performed. A number of simulations have been done in order to evaluate the performance of the designed system. The comparative results demonstrated the efficiency of using the T2FNN system in the diagnosis of diabetes. The physician can use the system for diabetes’ diagnosis.

A novel fuzzy community detection method based on improved community adjacency list

With the development of economy and society, network analysis is widely used in more and more fields. Signed network has a good effect in the process of representation and display. As an important part of network analysis, fuzzy community detection plays an increasingly important role in analyzing and visualizing the real world. Fuzzy community detection helps to detect nodes that belong to some communities but are still closely related to other communities. These nodes are helpful for mining information from the network more realistically. However, there is little research in this field. This paper proposes a fuzzy community detection algorithm based on pointer and adjacency list. The model adopts a new ICALF network data structure, which can achieve the effect of storing community partition structure and membership value between community and node at the same time, with low time complexity and storage space. Experiments on real networks verify the correctness of the method, and prove that the method is suitable for large-scale network applications.

Disentangling feature and lazy training in deep neural networks

Two distinct limits for deep learning have been derived as the network width h → ∞, depending on how the weights of the last layer scale with h. In the neural tangent Kernel (NTK) limit, the dynamics becomes linear in the weights and is described by a frozen kernel Θ (the NTK). By contrast, in the mean-field limit, the dynamics can be expressed in terms of the distribution of the parameters associated with a neuron, that follows a partial differential equation. In this work we consider deep networks where the weights in the last layer scale as αh −1/2 at initialization. By varying α and h, we probe the crossover between the two limits. We observe two the previously identified regimes of ‘lazy training’ and ‘feature training’. In the lazy-training regime, the dynamics is almost linear and the NTK barely changes after initialization. The feature-training regime includes the mean-field formulation as a limiting case and is characterized by a kernel that evolves in time, and thus learns some features. We perform numerical experiments on MNIST, Fashion-MNIST, EMNIST and CIFAR10 and consider various architectures. We find that: (i) the two regimes are separated by an α* that scales as . (ii) Network architecture and data structure play an important role in determining which regime is better: in our tests, fully-connected networks perform generally better in the lazy-training regime, unlike convolutional networks. (iii) In both regimes, the fluctuations δF induced on the learned function by initial conditions decay as , leading to a performance that increases with h. The same improvement can also be obtained at an intermediate width by ensemble-averaging several networks that are trained independently. (iv) In the feature-training regime we identify a time scale , such that for t ≪ t 1 the dynamics is linear. At t ∼ t 1, the output has grown by a magnitude and the changes of the tangent kernel | |ΔΘ| | become significant. Ultimately, it follows for ReLU and Softplus activation functions, with a < 2 and a → 2 as depth grows. We provide scaling arguments supporting these findings.

Analysing International Student Mobility Flows in Higher Education: A Comparative Study on European Countries

The analysis of the factors pulling and pushing students in a foreign country to complete their higher education is key for the implementation of university policies aimed at increasing the number of credits gained by students abroad and hence the degree of internationalisation of institutions. Consistent with previous studies, the present contribution examines the roles played by countries in the Erasmus student mobility flows by considering a joint strategy of analysis based on social network analysis and exploratory data analysis. First, data on Erasmus student exchanges among countries are gathered at macro-level from the European Union Open Data Portal and network data structures are analysed. Second, educational indicators from the Eurostat website are collected to describe the investments in higher education. The main findings suggest the presence of a core-periphery configuration in the student mobility network with few central countries in which the economic benefits and the investments in education seem to act as key elements for university attractiveness.

Efficient sentinel surveillance strategies for preventing epidemics on networks.

Surveillance plays a crucial role in preventing emerging infectious diseases from becoming epidemic. In circumstances where it is possible to monitor the infection status of certain people, transport hubs, or hospitals, early detection of the disease allows interventions to be implemented before most of the damage can occur, or at least its impact can be mitigated. This paper addresses the question of which nodes we should select in a network of individuals susceptible to some infectious disease in order to minimize the number of casualties. By simulating disease outbreaks on a collection of empirical and synthetic networks we show that the best strategy depends on topological characteristics of the network. For highly modular or spatially embedded networks it is better to place the sentinels on nodes distributed across different regions. However, if the degree heterogeneity is high, then a strategy that targets network hubs is preferred. We further consider the consequences of having an incomplete sample of the network and demonstrate that the value of new information diminishes as more data is collected. Finally we find further marginal improvements using two heuristics informed by known results in graph theory that exploit the fragmented structure of sparse network data.

Lossy asymptotic equipartition property for geometric networked data structures

This article extends the Generalized Asypmtotic Equipartition Property of Networked Data Structures to cover the Wireless Sensor Network modelled as coloured geometric random graph (CGRG). The main techniques used to prove this result remains large deviation principles for properly defined empirical measures on CGRGs. As a motivation for this article, we apply our results to some data from Wireless Sensor Network for Monitoring Water Quality from a Lake.