Search

Abstract As networks grow in size and complexity, backbones become an essential network representation. Indeed, they provide a simplified yet informative overview of the underlying organization by retaining the most significant and structurally influential connections within a network. Network heterogeneity often results in complex and intricate structures, making it challenging to identify the backbone. In response, we introduce the Multilevel Backbone Extraction Framework, a novel approach that diverges from conventional backbone methodologies. This generic approach prioritizes the mesoscopic organization of networks. First, it splits the network into homogeneous-density components. Second, it extracts independent backbones for each component using any classical Backbone technique. Finally, the various backbones are combined. This strategy effectively addresses the heterogeneity observed in network groupings. Empirical investigations on real-world networks underscore the efficacy of the Multilevel Backbone approach in preserving essential network structures and properties. Experiments demonstrate its superiority over classical methods in handling network heterogeneity and enhancing network integrity. The framework is adaptable to various types of networks and backbone extraction techniques, making it a versatile tool for network analysis and backbone extraction across diverse network applications.

Diabetes has emerged as a prevalent disease, affecting millions of individuals annually according to statistics. Numerous studies have delved into identifying key genes implicated in the causal mechanisms of diabetes. This paper specifically concentrates on 20 functional genes identified in various studies contributing to the complexities associated with Type 2 diabetes (T2D), encompassing complications such as nephropathy, retinopathy, cardiovascular disorders, and foot ulcers. These functional genes serve as a foundation for identifying regulatory genes, their regulators, and protein-protein interactions.The current study introduces a multi-layer Knowledge Graph (DbKB based on MSNMD: Multi-Scale Network Model for Diabetes), encompassing biological networks such as gene regulatory networks and protein-protein interaction networks. This Knowledge Graph facilitates the visualization and querying of inherent relationships between biological networks associated with diabetes, enabling the retrieval of regulatory genes, functional genes, interacting proteins, and their relationships.Through the integration of biologically relevant genetic, molecular, and regulatory information, we can scrutinize interactions among T2D candidate genes [1] and ascertain diseased genes [2]. The first layer of regulators comprises direct regulators to the functional genes, sourced from the TRRUST database in the human transcription factors dataset, thereby forming a multi-layered directed graph. A comprehensive exploration of these direct regulators reveals a total of 875 regulatory transcription factors, constituting the initial layer of regulating transcription factors. Moving to the second layer, we identify 550 regulatory genes.These functional genes engage with other proteins to form complexes, exhibiting specific functions. Leveraging these layers, we construct a Knowledge Graph aimed at identifying interaction-driven sub-networks involving (i) regulating functional genes, (ii) functional genes, and (iii) protein-protein interactions.

Abstract Networks are essential for analyzing complex systems. However, their growing size necessitates backbone extraction techniques aimed at reducing their size while retaining critical features. In practice, selecting, implementing, and evaluating the most suitable backbone extraction method may be challenging. This paper introduces netbone, a Python package designed for assessing the performance of backbone extraction techniques in weighted networks. Its comparison framework is the standout feature of netbone. Indeed, the tool incorporates state-of-the-art backbone extraction techniques. Furthermore, it provides a comprehensive suite of evaluation metrics allowing users to evaluate different backbones techniques. We illustrate the flexibility and effectiveness of netbone through the US air transportation network analysis. We compare the performance of different backbone extraction techniques using the evaluation metrics. We also show how users can integrate a new backbone extraction method into the comparison framework. netbone is publicly available as an open-source tool, ensuring its accessibility to researchers and practitioners. Promoting standardized evaluation practices contributes to the advancement of backbone extraction techniques and fosters reproducibility and comparability in research efforts. We anticipate that netbone will serve as a valuable resource for researchers and practitioners enabling them to make informed decisions when selecting backbone extraction techniques to gain insights into the structural and functional properties of complex systems.

Recent advancements in 3D modeling have revolutionized various fields, including virtual reality, computer-aided diagnosis, and architectural design, emphasizing the importance of accurate quality assessment for 3D point clouds. As these models undergo operations such as simplification and compression, introducing distortions can significantly impact their visual quality. There is a growing need for reliable and efficient objective quality evaluation methods to address this challenge. In this context, this paper introduces a novel methodology to assess the quality of 3D point clouds using a deep learning-based no-reference (NR) method. First, it extracts geometric and perceptual attributes from distorted point clouds and represent them as a set of 1D vectors. Then, transfer learning is applied to obtain high-level features using a 1D convolutional neural network (1D CNN) adapted from 2D CNN models through weight conversion from ImageNet. Finally, quality scores are predicted through regression utilizing fully connected layers. The effectiveness of the proposed approach is evaluated across diverse datasets, including the Colored Point Cloud Quality Assessment Database (SJTU_PCQA), the Waterloo Point Cloud Assessment Database (WPC), and the Colored Point Cloud Quality Assessment Database featured at ICIP2020. The outcomes reveal superior performance compared to several competing methodologies, as evidenced by enhanced correlation with average opinion scores.

Theoretical and empirical studies on diffusion models have revealed their versatile applicability across different fields, spanning from sociology and finance to biology and ecology. The presence of a community structure within real-world networks has a substantial impact on how diffusion processes unfold. Key nodes located both within and between these communities play a crucial role in initiating diffusion, and community-aware centrality measures effectively identify these nodes. While numerous diffusion models have been proposed in literature, very few studies investigate the relationship between the diffusive ability of key nodes selected by community-aware centrality measures, the distinct dynamical conditions of various models, and the diverse network topologies. By conducting a comparative evaluation across four diffusion models, utilizing both synthetic and real-world networks, along with employing two different community detection techniques, our study aims to gain deeper insights into the effectiveness and applicability of the community-aware centrality measures. Results suggest that the diffusive power of the selected nodes is affected by three main factors: the strength of the network's community structure, the internal dynamics of each diffusion model, and the budget availability. Specifically, within the category of simple contagion models, such as SI, SIR, and IC, we observe similar diffusion patterns when the network's community structure strength and budget remain constant. In contrast, the LT model, which falls under the category of complex contagion dynamics, exhibits divergent behavior under the same conditions.

Graph distance measures have emerged as an effective tool for evaluating the similarity or dissimilarity between graphs. Recently, there has been a growing trend in the application of movie networks to analyze and understand movie stories. Previous studies focused on computing the distance between individual characters in narratives and identifying the most important ones. Unlike previous techniques, which often relied on representing movie stories through single-layer networks based on characters or keywords, a new multilayer network model was developed to allow a more comprehensive representation of movie stories, including character, keyword, and location aspects. To assess the similarities among movie stories, we propose a methodology that utilizes a multilayer network model and layer-to-layer distance measures. We aim to quantify the similarity between movie networks by verifying two aspects: (i) regarding many components of the movie story and (ii) quantifying the distance between their corresponding movie networks. We tend to explore how five graph distance measures reveal the similarity between movie stories in two aspects: (i) finding the order of similarity among movies within the same genre, and (ii) classifying movie stories based on genre. We select movies from various genres: sci-fi, horror, romance, and comedy. We extract movie stories from movie scripts regarding character, keyword, and location entities to perform this. Then, we compute the distance between movie networks using different methods, such as the network portrait divergence, the network Laplacian spectra descriptor (NetLSD), the network embedding as matrix factorization (NetMF), the Laplacian spectra, and D-measure. The study shows the effectiveness of different methods for identifying similarities among various genres and classifying movies across different genres. The results suggest that the efficiency of an approach on a specific network type depends on its capacity to capture the inherent network structure of that type. We propose incorporating the approach into movie recommendation systems.

Medical diagnosis research has recently focused on feature selection techniques due to the availability of multiple variables in medical datasets. Wrapper-based feature selection approaches have shown promise in providing faster and more cost-effective predictors. However, selecting the most relevant features from medical datasets to increase disease classification accuracy remains a challenging research issue. To address this challenge, we propose an effective wrapper-based feature selection approach called BTLBO-KNN. It combines an improved Binary Teaching-Learning Based Optimization (BTLBO) algorithm with the K-Nearest Neighbor (KNN) classifier to accelerate the convergence rate in finding the near-optimal features subset. BTLBO-KNN incorporates two new efficient binary teaching and learning processes, an abandoned learner’s replacement mechanism, and a teacher knowledge improvement method. We extensively compare BTLBO-KNN with recent state-of-the-art wrapper-based feature selection approaches on COVID-19 and 23 gene-expression and medical datasets with different dimensional complexities. Our results demonstrate the superiority of BTLBO-KNN over its alternatives in terms of minimizing the number of selected features and the classification error rate.

Meningitis, characterized by meninges and cerebrospinal fluid inflammation, poses diagnostic challenges due to diverse clinical manifestations. This work introduces an explainable AI automatic medical decision methodology that determines critical features and their relevant values for the differential diagnosis of various meningitis cases. We proceed with knowledge acquisition to define the rules for this research. Currently, we have established the etiological diagnosis of Meningococcaemia, Meningococcal Meningitis, Tuberculous Meningitis, Aseptic Meningitis, Haemophilus influenzae Meningitis, and Pneumococcal Meningitis. The data preprocessing was conducted after collecting data from samples with meningitis diseases at Setif Hospital in Algeria. Tree-based ensemble methods were then applied to assess the model’s performance. Finally, we implement an XAI agnostic explainability approach based on the SHapley Additive exPlanations technique to attribute each feature’s contribution to the model’s output. Experiments were conducted on the collected dataset and the SINAN database, obtained from the Brazilian Government’s Health Information System on Notifiable Diseases, which comprises 6729 patients aged over 18 years. The Extreme Gradient Boosting model was chosen for its superior performance metrics (Accuracy: 0.90, AUROC: 0.94, and F1-score: 0.98). Setif’s hospital data revealed notable performance metrics (Accuracy: 0.7143, F1-Score: 0.7857). This study’s findings showcase each feature’s contribution to the model’s predictions and diagnosis. It also reveals critical biomarker ranges associated with distinct types of Meningitis. Significant diagnostic effect was found for Meningococcal Meningitis with elevated neutrophil levels ( $ \gt $ 40%) and balanced lymphocyte levels (40%–60%). Tuberculous Meningitis demonstrated low neutrophil levels ( $ \lt $ 60%) and elevated lymphocyte levels ( $ \gt $ 60%). H. influenzae meningitis exhibited a predominance of neutrophils ( $ \gt $ 80%), while Aseptic meningitis showed lower neutrophil levels ( $ \lt $ 40%) and lymphocyte levels within the range of 50%–60%. The majority of the AI automatic medical decision results are twinned with validation by our team of infectious disease experts, confirming the alignment of algorithmic diagnoses with clinical practices.

Mobility is a critical element for understanding human contact networks. In many studies, the researchers use random processes to model human mobility. However, people do not move randomly in their environment. Their interactions do not depend only on spatial constraints but on their temporal, social, economic, and cultural activities. The topological structure of the physical and/or proximity contact networks depends, therefore, entirely on the mobility patterns. This paper performs an extensive comparative analysis of real-world temporal contact networks and synthetic networks based on influential mobility models. Results show that the various topological properties of most of the synthetic datasets depart from those observed in real-world contact networks because the randomness of some mobility parameters tends to move away from human contact properties. However, it appears that data generated using Spatio-Temporal Parametric Stepping (STEPS) mobility model reveals similarities with real temporal contact networks such as heavy-tailed distribution of contact duration, frequency of pairs of contacts, and the bursty phenomenon. These results pave the way for further improvement of mobility models to generate meaningful artificial contact networks.

The topological structure of the world air transportation network has been the subject of much research. However, to better understand the reality of air networks, one can consider the traffic, the number of passengers, or the distance between flights. This paper studies the weighted world air transportation network through the component structure, recently introduced in the network literature, by using the number of flights. The component structure is based on the community or multiple core-periphery structures and splits the network into local and global components. The local components capture the regional flights of these two mesoscopic structures (dense parts). The global components capture the inter-regional flights (links between the dense parts). We perform a comparative analysis of the world air transportation network and its components with their weighted counterparts. Moreover, we explore the strength and the s-core of these networks. Results display fewer local components well delimited and more global components covering the world than the unweighted world air transportation network. Centrality analysis reveals the difference between the top airports with high traffic and the top airports with high degrees. This difference is more pronounced in the global air network and the largest global component. Core analysis shows similitude between the s-core and the k-core for the local and global components, even though the latter includes more airports. For the world air network, the North and Central America-Caribbean airports dominate the s-core, whereas the European airports dominate the k-core.

Abstract Air transportation plays an essential role in the global economy. Therefore, there is a great deal of work to understand better the complex network formed by the links between the origins and destinations of flights. Some investigations show that the world air transportation network exhibits a community and a core-periphery structure. Although precious, these representations do not distinguish the inter-regional (global) web of connections from the regional (local) one. Therefore, we propose a new mesoscopic model called the component structure that decomposes the network into local and global components. Local components are the dense areas of the network, and global components are the nodes and links bridging the local components. As a case study, we consider the unweighted and undirected world air transportation network. Experiments show that it contains seven large local components and multiple small ones spatially well-defined. Moreover, it has a main global component covering the world. We perform an extensive comparative analysis of the structure of the components. Results demonstrate the non-homogeneous nature of the world air transportation network. The local components structure highlights regional differences, and the global component organization captures the efficiency of inter-regional travel. Centrality analysis of the components allows distinguishing airports centered on regional destinations from those focused on inter-regional exchanges. Core analysis is more accurate in the components than in the whole network where Europe dominates, blurring the rest of the world. Besides the world air transportation network, this paper demonstrates the potential of the component decomposition for modeling and analyzing the mesoscale structure of networks.

Abstract Identifying vital nodes in networks exhibiting a community structure is a fundamental issue. Indeed, community structure is one of the main properties of real-world networks. Recent works have shown that community-aware centrality measures compare favorably with classical measures agnostic about this ubiquitous property. Nonetheless, there is no clear consensus about how they relate and in which situation it is better to use a classical or a community-aware centrality measure. To this end, in this paper, we perform an extensive investigation to get a better understanding of the relationship between classical and community-aware centrality measures reported in the literature. Experiments use artificial networks with controlled community structure properties and a large sample of real-world networks originating from various domains. Results indicate that the stronger the community structure, the more appropriate the community-aware centrality measures. Furthermore, variations of the degree and community size distribution parameters do not affect the results. Finally, network transitivity and community structure strength are the most significant drivers controlling the interactions between classical and community-aware centrality measures.

This paper proposes a new method for blind mesh visual quality assessment (MVQA) based on a graph convolutional network. For that, we address the node classification problem to predict the perceived visual quality. First, two matrices representing the 3D mesh are considered: a graph adjacency matrix and a feature matrix. Both matrices are used as input to a shallow graph convolutional network. The network consists of two convolutional layers followed by a max-pooling layer to provide the final feature representation. With this structure, the Softmax classifier predicts the quality score category without the reference mesh’s availability. Experiments are conducted on four publicly available databases constructed explicitly for the mesh quality assessment task. We investigate several perceptual and visual features to select the most effective combination. Comparisons with the state-of-the-art alternative methods show the effectiveness of the proposed framework.

In COVID-19 related infodemic, social media becomes a medium for wrongdoers to spread rumors, fake news, hoaxes, conspiracies, astroturf memes, clickbait, satire, smear campaigns, and other forms of deception. It puts a tremendous strain on society by damaging reputation, public trust, freedom of expression, journalism, justice, truth, and democracy. Therefore, it is of paramount importance to detect and contain unreliable information. Multiple techniques have been proposed to detect fake news propagation in tweets based on tweets content, propagation on the network of users, and the profile of the news generators. Generating human-like content allows deceiving content-based methods. Network-based methods rely on the complete graph to detect fake news, resulting in late detection. User profile-based techniques are effective for bots or fake accounts detection. However, they are not suited to detect fake news from original accounts. To deal with the shortcomings in existing methods, we introduce a source-based method focusing on the news propagators’ community, including posters and re-tweeters to detect such contents. Propagators are connected using follower-following relations. A feature set combining the connectivity patterns of news propagators with their profile features is used in a machine learning framework to perform binary classification of tweets. Complex network measures and user profile features are also examined separately. We perform an extensive comparative analysis of the proposed methodology on a real-world COVID-19 dataset, exploiting various machine learning and deep learning models at the community and node levels. Results show that hybrid features perform better than network features and user features alone. Further optimization demonstrates that Ensemble’s boosting model CATBoost and deep learning model RNN are the most effective, with an AUC score of 98%. Furthermore, preliminary results show that the proposed solution can also handle fake news in the political and entertainment domain using a small training set.

While most of opinion formation models consider static networks, a dynamic opinion formation model is proposed in this work. The so-called Temporal Threshold Page Rank Opinion Formation model (TTPROF) integrates temporal evolution in two ways. First, the opinion of the agents evolves with time. Second, the network structure is also time varying. More precisely, the relations between agents evolve with time. In the TTPROF model, a node is affected by the part of its neighbor's opinions weighted by their Page Rank values. A threshold is introduced to limit the neighbors that can share their opinion. In other words, a neighbor influences a node if the difference between their opinions is below the threshold. Finally, a fraction of top ranked nodes in the neighborhood is considered influential nodes irrespective of the threshold value. Experiments have been performed on random temporal networks to analyze how opinions propagate and converge to consensus or multiple clusters. Preliminary results have been presented. In this paper, this work is extended in two directions. First, the impact of various centrality measures on model behavior is investigated. Indeed, in earlier work, the influence of a node is measured using Page Rank. New results using Directed Degree centrality and Closeness Centrality are derived. They allow to compare global against local influence measures as well as distance-based centrality and to better understand the impact of the weighting parameter on the model convergence. Second, the results of an extensive experimental investigation are reported and analyzed in order to characterize the model convergence in various situations.

In real-world networks, nodes are usually organized into modules or communities of densely connected nodes. In situations where nodes can belong to multiple communities we say that the communities overlap, and the nodes shared by more than one community are called the overlapping nodes. This occurs especially in social networks where an individual belongs to various social groups and organizations such as working circles, family, friendship or virtual groups on the Internet. Complex networks are known to have a heavy tail degree distribution. Indeed, they are organized with a vast majority of nodes with few interactions and a small set of highly connected nodes called hubs. In this paper, our goal is to study the relationship between the overlapping nodes and the hubs. Indeed, we suspect that the hubs are in the vicinity of the overlapping nodes. If this assumption is confirmed, it gives a new perspective on how the communities are organized and of the crucial importance of the overlapping nodes. In an attempt to investigate the ubiquity of this property, we perform series of experiments on various real-world networks with overlapping community structure. Results show that the hubs represent always a large proportion of the one-step neighbors of overlapping nodes. These results may have implications in various contexts. For example, searching for the hubs in large networks can be done starting from the overlapping nodes. Furthermore, this study may also provide new directions for designing new community detection algorithms.

Vector-borne diseases such as malaria, dengue fever, West Nile virus, and so forth are some of the most prominent threats to human health. They are transmitted to the human population by infected insects or by direct transmission between humans. The epidemic process relies on suitable environmental and climatic conditions. Indeed, climatic factors affect the development of pathogens in vectors as well as the population dynamics of the vectors impacting significantly the incidence of disease in the human population. While the influence of the climatic conditions on Vector-borne diseases is well-documented, there is a strong need to design more realistic epidemiological models incorporating environmental features that show a close relationship with the epidemic process observed in the human population. Indeed, classical models concentrate either on the climatic influence on the vector population dynamics or on the connectivity patterns of the host population, loosing the full picture of the epidemic process dynamics. Inspired by real data of infectious diseases, a Seasonal Susceptible-Infected-Recovered (Seasonal SIR) epidemiological model is developed and analyzed. The proposed model incorporates the influence of the temperature variations together with the heterogeneous structure of the human interaction network on the spreading process of vector-borne diseases. Simulations are performed in order to get a better understanding of the climate variations and of the heterogeneous nature of the contact network on the transmission dynamics. Results show that failing to incorporate these features on the model can lead to a poor estimation of the maximum fraction of infected individuals in the host population. Furthermore there is a serious influence on the time needed to reach this maximum. The Seasonal SIR model proves to finely model the dynamics of outbreaks observed in real-world situations. It provides a basis for more effective predictions of disease outbreaks that can be used in order to implement appropriate control measures to contain epidemics.

Recommender systems are information software that retrieves relevant items for users from massive sources of data. The variational autoencoder (VAE) has proven to be a promising approach for recommendation systems, as it can explore high-level user-item relations and extract contingencies from the input effectively. However, the previous variants of VAE have so far seen limited application to domain-specific recommendations that require additional side information. Hence, The Ensemble Variational Autoencoder framework for recommendations (EnsVAE) is proposed. This architecture specifies a procedure to transform sub-recommenders' predicted utility matrix into interest probabilities that allow the VAE to represent the variation in their aggregation. To evaluate the performance of EnsVAE, an instance - called the “Ensemblist GRU/GLOVE model” - is developed. It is based on two innovative recommender systems: 1-) a new “GloVe content-based filtering recommender” (GloVe-CBF) that exploits the strengths of embedding-based representations and stacking ensemble learning techniques to extract features from the item-based side information. 2-) a variant of neural collaborative filtering recommender, named “Gate Recurrent Unit-based Matrix Factorization recommender” (GRU-MF). It models a high level of non-linearities and exhibits interactions between users and items in latent embeddings, reducing user biases towards items that are rated frequently by users. The developed instance speeds up the reconstruction of the utility matrix with increased accuracy. Additionally, it can switch between one of its sub-recommenders according to the context of their use. Our findings reveal that EnsVAE instances retain as much information as possible during the reconstruction of the utility matrix. Furthermore, the trained VAE's generative trait tackles the cold-start problem by accurately estimating the interest probabilities of newly-introduced users and resources. The empirical study on real-world datasets proves that EnsVAE significantly outperforms the state-of-the-art methods in terms of recommendation performances.

Hierarchy and centrality are two popular notions used to characterize the importance of entities in complex systems. Indeed, many complex systems exhibit a natural hierarchical structure, and centrality is a fundamental characteristic allowing to identify key constituents. Several measures based on various aspects of network topology have been proposed in order to quantify these concepts. While numerous studies have investigated whether centrality measures convey redundant information, how centrality and hierarchy measures are related is still an open issue. In this paper, we investigate the association between centrality and hierarchy using several correlation and similarity evaluation measures. A series of experiments is performed in order to evaluate the combinations of 6 centrality measures with 4 hierarchy measures across 28 diverse real-world networks with varying topological characteristics. Results show that network density and transitivity play a key role in shaping the redundancy between centrality and hierarchy measures.

The robustness of a system indicates its ability to withstand disturbances while maintaining its properties, performance, and efficiency. There are plenty of studies on the robustness of air transport networks in the literature. However, few works consider its mesoscopic organization. Building on the recently introduced component structure, we explore the impact of targeted attacks on the weighted world air transportation network on its components. Indeed, it contains five local components covering different regions (North America-Caribbean, Europe-Russia, East and Southeast Asia-Oceania, Africa-Middle East-Southern Asia, and South America) and one global component linking these regions. We investigate targeted attacks based on influential weighted centrality measures (strength, betweenness, and PageRank). Results show that the local components gradually separate from the world air transportation network as the fraction of removed airports grows. The weighted betweenness attack removes fewer top airports to isolate the regions compared to its alternatives. Furthermore, it is still convenient to travel locally in the separate areas. In contrast, strength and PageRank attacks need to target more airports to split the network. However, they are more disruptive. Indeed, the size of the isolated local components reduces drastically, so it becomes more challenging to travel locally. Looking at the world air transportation network through its component structure reveals a new viewpoint on its resilence. It opens new perspectives to design more efficient attacks.

The main objective of Zika transmission studies is to work out the simplest approach to scale back human mortality and morbidity caused by the disease. Therefore, it is essential to spot the relative importance of the various factors contributing to the transmission and prevalence of the disease. Many mathematical models have been formulated incorporating vector-to-human transmission or human-to-human transmission. However, they do not take into consideration the mixture of both sorts of transmission. It raises the question of the impact of both sorts of transmission on the disease dynamics. We develop a mathematical model of Zika with the vertical transmission in the vector population and human-to-human transmission to answer this question. It includes the immature phase of mosquitoes (eggs), adult mosquitoes (susceptible, exposed, and infectious), and human hosts (susceptible, exposed, infectious, and recovered). Results show that neglecting sexual transmission results in an understatement of the proportion of the infected population. Furthermore, it reduces the speed of disease spreading. On the other hand, vertical transmission in mosquitoes has a negligible effect on the dynamics of disease spread. We perform a sensitivity analysis of the reproductive number R0 to raise to understand the parameters driving the dynamics of the disease. It appears that the most sensitive parameters in decreasing order are as follows: the adult mosquito death rate, the sting rate, the transmission probability of mosquito to human, and, therefore, the transmission probability of human to mosquito. Furthermore, the proportion at the equilibrium of infected humans is extremely sensitive to the transition rate from the immature vector stage to the adult stage, the human-to-human transmission rate, and, therefore, the human recovery rate. These results confirm that control policies targeting the vector population and, therefore, the recovery rate of people are pretty effective solutions. To validate the model and estimate the important parameters of the model and the prediction of the disease, we consider the real cases in Colombia from 2016. In a series of graphic maps, we presented the comparative study to estimate the disease scenarios and to predict the time limit of the epidemic control measure.

Quantifying a node's importance is decisive for developing efficient strategies to curb or accelerate any spreading phenomena. Centrality measures are well-known methods used to quantify the influence of nodes by extracting information from the network's structure. The pitfall of these measures is to pinpoint nodes located in the vicinity of each other, saturating their shared zone of influence. In this paper, we propose a ranking strategy exploiting the ubiquity of the community structure in real-world networks. The proposed community-aware ranking strategy naturally selects a set of distant spreaders with the most significant influence in the networks. One can use it with any centrality measure. We investigate its effectiveness using real-world and synthetic networks with controlled parameters in a Susceptible-Infected-Recovered (SIR) diffusion model scenario. Experimental results indicate the superiority of the proposed ranking strategy over all its counterparts agnostic about the community structure. Additionally, results show that it performs better in networks with a strong community structure and a high number of communities of heterogeneous sizes.

Abstract Community structure is one of the most relevant features encountered in numerous real-world applications of networked systems. Despite the tremendous effort of a large interdisciplinary community of scientists working on this subject over the past few decades to characterize, model, and analyze communities, more investigations are needed in order to better understand the impact of community structure and its dynamics on networked systems. Here, we first focus on generative models of communities in complex networks and their role in developing strong foundation for community detection algorithms. We discuss modularity and the use of modularity maximization as the basis for community detection. Then, we follow with an overview of the Stochastic Block Model and its different variants as well as inference of community structures from such models. Next, we focus on time evolving networks, where existing nodes and links can disappear, and in parallel new nodes and links may be introduced. The extraction of communities under such circumstances poses an interesting and non-trivial problem that has gained considerable interest over the last decade. We briefly discuss considerable advances made in this field recently. Finally, we focus on immunization strategies essential for targeting the influential spreaders of epidemics in modular networks. Their main goal is to select and immunize a small proportion of individuals from the whole network to control the diffusion process. Various strategies have emerged over the years suggesting different ways to immunize nodes in networks with overlapping and non-overlapping community structure. We first discuss stochastic strategies that require little or no information about the network topology at the expense of their performance. Then, we introduce deterministic strategies that have proven to be very efficient in controlling the epidemic outbreaks, but require complete knowledge of the network.

Abstract Discovering content and stories in movies is one of the most important concepts in multimedia content research studies. Network models have proven to be an efficient choice for this purpose. When an audience watches a movie, they usually compare the characters and the relationships between them. For this reason, most of the modelsdeveloped so far are based on social networks analysis. They focus essentially on the characters at play. By analyzing characters interactions, we can obtain a broad picture of the narration’s content. Other works have proposed to exploit semantic elements such as scenes, dialogues, etc.. However, they are always captured from a single facet. Motivated by these limitations, we introduce in this work a multilayer network model to capture the narration of a movie based on its script, its subtitles, and the movie content. After introducing the model and the extraction process from the raw data, weperform a comparative analysis of the whole 6-movie cycle of the Star Wars saga. Results demonstrate the effectiveness of the proposed framework for video content representation and analysis.

Many web platforms now include recommender systems. Network representation learning has been a successful approach for building these efficient recommender systems. However, learning the mutual influence of nodes in the network is challenging. Indeed, it carries collaborative signals accounting for complex user-item interactions on user decisions. For this purpose, in this paper, we develop a Mutual Interaction Graph Attention Network “MIGAN”, a new algorithm based on self-supervised representation learning on a large-scale bipartite graph (BGNN). Experimental investigation with real-world data demonstrates that MIGAN compares favorably with the baselines in terms of prediction accuracy and recommendation efficiency.

Abstract Identifying influential spreaders in networks is an essential issue in order to prevent epidemic spreading, or to accelerate information diffusion. Several centrality measures take advantage of various network topological properties to quantify the notion of influence. However, the vast majority of works ignore its community structure while it is one of the main features of many real-world networks. In a recent study, we show that the centrality of a node in a network with non-overlapping communities depends on two features: Its local influence on the nodes belonging to its community, and its global influence on the nodes belonging to the other communities. Using global and local connectivity of the nodes, we introduced a framework allowing to redefine all the classical centrality measures (designed for networks without community structure) to non-overlapping modular networks. In this paper, we extend the so-called “Modular Centrality” to networks with overlapping communities. Indeed, it is a frequent scenario in real-world networks, especially for social networks where nodes usually belong to several communities. The “Overlapping Modular Centrality” is a two-dimensional measure that quantifies the local and global influence of overlapping and non-overlapping nodes. Extensive experiments have been performed on synthetic and real-world data using the Susceptible-Infected-Recovered (SIR) epidemic model. Results show that the Overlapping Modular Centrality outperforms its alternatives designed for non-modular networks. These investigations provide better knowledge on the influence of the various parameters governing the overlapping community structure on the nodes’ centrality. Additionally, two combinations of the components of the Overlapping Modular Centrality are evaluated. Comparative analysis with competing methods shows that they produce more efficient centrality scores.

Abstract Identifying influential nodes in a network is a fundamental issue due to its wide applications, such as accelerating information diffusion or halting virus spreading. Many measures based on the network topology have emerged over the years to identify influential nodes such as Betweenness, Closeness, and Eigenvalue centrality. However, although most real-world networks are made of groups of tightly connected nodes which are sparsely connected with the rest of the network in a so-called modular structure, few measures exploit this property. Recent works have shown that it has a significant effect on the dynamics of networks. In a modular network, a node has two types of influence: a local influence (on the nodes of its community) through its intra-community links and a global influence (on the nodes in other communities) through its inter-community links. Depending on the strength of the community structure, these two components are more or less influential. Based on this idea, we propose to extend all the standard centrality measures defined for networks with no community structure to modular networks. The so-called “Modular centrality” is a two-dimensional vector. Its first component quantifies the local influence of a node in its community while the second component quantifies its global influence on the other communities of the network. In order to illustrate the effectiveness of the Modular centrality extensions, comparison with their scalar counterparts is performed in an epidemic process setting. Simulation results using the Susceptible-Infected-Recovered (SIR) model on synthetic networks with controlled community structure allows getting a clear idea about the relation between the strength of the community structure and the major type of influence (global/local). Furthermore, experiments on real-world networks demonstrate the merit of this approach.

News reports in media contain news about society’s social and political conditions. With the help of publicly available digital datasets of events, it is possible to study a complex network of mass violations, i.e., Mass Killings. Multiple approaches have been applied to bring essential insights into the events and involved actors. Power law distribution behavior finds in the tail of actor mention, co-actor mention, and actor degree tells us about the dominant behavior of influential actors that grows their network with time. The United States, France, Israel, and a few other countries have been identified as major players in the propagation of Mass Killing throughout the past 20 years. It is demonstrated that targeting the removal of influential actors may stop the spreading of such conflicting events and help policymakers and organizations. This paper aims to identify and formulate the conflicts with the actor’s perspective at a global level for a period of time. This process is a generalization to be applied to any level of news, i.e., it is not restricted to only the global level.

Herding has often been blamed as one of the possible causes of market instabilities, ultimately yielding to bubbles and crushes. On the other hand, researchers hypothesized that financial systems may benefit from the so-called wisdom of the crowd. To solve this apparent dichotomy, we leverage a novel financial market model, where the agents form their expectations by combining their individual return estimation with the expectations of their neighbors. By establishing a link between herding, sociality, and market instabilities, we point out that the emergence of collective decisions in the market is not necessarily detrimental. Indeed, when all the agents tend to conform their expectations to those of one or few leaders, herding might dramatically reduce market efficiency. However, when each agent accounts for a plurality of opinions, thus following the wisdom of the crowd, market dynamics become efficient. Following these observations, we propose two alternative control strategies to reduce market instability and enhance its efficiency.

In this paper, we study the topological properties of the global supply chain network in terms of its degree distribution, clustering coefficient, degree-degree correlation, bow-tie structure, and community structure to test the efficient supply chain propositions proposed by E. J.S. Hearnshaw et al. The global supply chain data in the year 2017 are constructed by collecting various company data from the web site of Standard & Poor’s Capital IQ platform. The in- and out-degree distributions are characterized by a power law of the form of γin = 2.42 and γout = 2.11. The clustering coefficient decays

In this paper, a robust hybrid watermarking method based on discrete wavelet transform (DWT), discrete cosine transform (DCT), and scale-invariant feature transformation (SIFT) is proposed. Indeed, it is of prime interest to develop robust feature-based image watermarking schemes to withstand both image processing attacks and geometric distortions while preserving good imperceptibility. To this end, a robust watermark is embedded in the DWT-DCT domain to withstand image processing manipulations, while SIFT is used to protect the watermark from geometric attacks. First, the watermark is embedded in the middle band of the discrete cosine transform (DCT) coefficients of the HL1 band of the discrete wavelet transform (DWT). Then, the SIFT feature points are registered to be used in the extraction process to correct the geometric transformations. Extensive experiments have been conducted to assess the effectiveness of the proposed scheme. The results demonstrate its high robustness against standard image processing attacks and geometric manipulations while preserving a high imperceptibility. Furthermore, it compares favorably with alternative methods.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Three-dimensional models have been extensively used in several applications including computer-aided design (CAD), video games, medical imaging due to the processing capability improvement of computers, and the development of network bandwidth. Therefore, the necessity of implementing 3D mesh watermarking schemes aiming to protect copyright has increased considerably. In this paper, a blind robust 3D mesh watermarking method based on mesh saliency and wavelet transform for copyright protection is proposed. The watermark is inserted by quantifying the wavelet coefficients using quantization index modulation (QIM) according to the mesh saliency of the 3D semiregular mesh. The synchronizing primitive is the distance between the mesh center and salient points in the descending order. The experimental results show the high imperceptibility of the proposed scheme while ensuring a good robustness against a wide range of attacks including smoothing, additive noise, element reordering, similarity transformations, etc.