With the rapid development of the cryptocurrency market, price volatility and liquidity have become key indicators for understanding market behavior. However, their joint dynamics are often shaped by higher-order, multi-attribute interactions that are not well captured by conventional low-order analyses. Existing approaches largely emphasize pairwise relationships and thus struggle to provide a system-level and time-varying characterization of whether multiple market attributes convey redundant overlap or synergistic complementarity, leaving an important gap in understanding multivariate market organization. This paper proposes a multi-scale network framework based on information dynamics, built on O-information theory and quantified through three parallel measures: global O-information rate (global OIR), local OIR, and the OIR gradient, to characterize redundancy–synergy structures in multi-attribute cryptocurrency systems. Using Binance Vision spot-market data for twelve major cryptocurrencies from August 1, 2020 to November 30, 2025 across four sampling intervals (2 h, 6 h, 12 h, and 1d), we conduct rolling-window estimation and systematic cross-scale comparisons. The results reveal pronounced cross-asset heterogeneity and support a distinction between leading and non-leading assets, with BTC, ETH, XRP, and BNB identified as leading assets while also exhibiting within-group differences. The findings provide a holistic and dynamic perspective on higher-order information organization in digital-asset markets and suggest that HOI-based indicators offer information that complements conventional connectedness measures in financial research.

We present a novel approach for assessing the reliability of error-prone systems and networks using an interpretable prototype-based classification framework. Specifically, reliability levels of consecutive k-out-of-n success systems, failure networks, and domination networks are classified using Generalized Matrix Learning Vector Quantization (GMLVQ). Beyond achieving accurate classification, the proposed method provides informative insights into the influence of input probabilities on the resulting reliability levels. To further enhance the interpretability of the learned relevance matrix in GMLVQ, we introduce two graph-based visualization strategies that reveal structural patterns and relevant feature interactions that are critical for reliability-level classification. The proposed approach is generally applicable to arbitrary coherent systems and can be further adapted to estimate the probability of a union of any finite family of events based on their individual and pairwise intersection probabilities.

The syntactic structure of a sentence can be described as a tree that indicates the syntactic relationships between words. In spite of significant progress in unsupervised methods that retrieve the syntactic structure of sentences, guessing the right direction of edges is still a challenge. As in a syntactic dependency structure edges are oriented away from the root, the challenge of guessing the right direction can be reduced to finding an undirected tree and the root. The limited performance of current unsupervised methods demonstrates the lack of a proper understanding of what a root vertex is from first principles. We consider an ensemble of centrality scores, some that only take into account the free tree (non-spatial scores) and others that take into account the position of vertices (spatial scores). We test the hypothesis that the root vertex is an important or central vertex of the syntactic dependency structure. We confirm the hypothesis in the sense that root vertices tend to have high centrality and that vertices of high centrality tend to be roots. The best performance in guessing the root is achieved by novel scores that only take into account the position of a vertex and that of its neighbours. We provide theoretical and empirical foundations towards a universal notion of rootness from a network science perspective.

Social networks have become one of the main structures for representing people’s relationships, so building a structure that accurately describes relationships is an essential issue for the scientific community focused on analyzing relationships. Currently, the construction of social networks is a research problem due to the complexity of measuring how connected people are to each other. In our work, we propose a social network construction approach to represent social relationships based on facial analysis from a collection of images or video frames, called FA-SNet. Instead of measuring the connectedness strength of relationships based on the frequency of joint appearances present in a collection, we measure how connected people are; that is, we provide a connectivity score that represents the strength or closeness of relationships. This connectivity information is obtained from a connectivity matrix calculated by measuring factors based on facial analysis, such as ‘co-occurrence’, ‘closeness’, ‘connection’, ‘empathy’, and ‘happiness’. The social network is composed of nodes interconnected by edges, where the nodes represent the individuals and the edges express their relationships. The key idea of the obtained graph is that the closer are the nodes, the higher the connectivity between them. We evaluated the proposed approach through four experiments using different datasets, such as two summary clips from the television series Pride and Prejudice, a video from the film Les Misérables, a summary clip from the film Emma, and a set of images from a private wedding celebration. The effectiveness of the constructed networks was measured using the Graph Edit Distance metric, which shows how similar the proposed network is to reference networks. The results obtained from the proposed approach show potential in identifying relationships between individuals and quantifying the strength of these relationships using connectivity information. We believe our methodology will be a contribution for future studies in relationship analysis. The project page can be found in: https://github.com/Toscanojasiel/FA-SNet_Social_Network_Construction

Abstract Knowledge graphs (KGs) are useful for analyzing social structures, community dynamics, institutional memberships, and other complex relationships across domains from sociology to public health. While recent advances in large language models (LLMs) have improved the scalability and accessibility of automated KG extraction from large text corpora, the impacts of extraction errors on downstream analyses are poorly understood, especially for applied scientists who depend on accurate KGs for real-world insights. To address this gap, we conducted the first evaluation of KG extraction performance at two levels: (1) micro-level edge accuracy, which is consistent with standard NLP evaluations, and manual identification of common error sources; (2) macro-level graph metrics that assess structural properties such as community detection and connectivity, which are relevant to real-world applications. Focusing on affiliation graphs of person membership in organizations extracted from social register books, our study identifies a range of extraction performance where biases across most downstream graph analysis metrics are near zero. However, as extraction performance declines, we find that many metrics exhibit increasingly pronounced biases, with each metric tending toward a consistent direction of either over- or under-estimation. Through simulations, we further show that error models commonly used in the literature do not capture these bias patterns, indicating the need for more realistic error models for KG extraction. Our findings provide actionable insights for practitioners and underscore the importance of advancing extraction methods and error modeling to ensure reliable and meaningful downstream analyses.

Cancer care relies on effective coordination within a multidisciplinary care team. Changes to teams due to departures remain understudied despite rising oncologist turnover in the United States. In this study, we aimed to investigate the impact of oncologist departures on the remaining care team members. We used Medicare claims associated with beneficiaries aged 66–99 to identify physicians involved in care for common cancer types (i.e., breast, lung, and colorectal cancer). We restricted our analysis to medical oncologists, radiation oncologists, and surgeons specializing in oncology (collectively, “oncologists”). We identified oncologists who left a practice location in 2017–2019 using the Medicare Carrier file and linked them to retained oncologists based on shared patients. Multivariable hierarchical linear regression was used to investigate how retained oncologists’ patient-sharing patterns changed after a colleague’s departure. Our results support that retained rural-practicing oncologists experienced an expansion and restructuring of their patient-sharing ties following oncologist departures while retained urban-practicing oncologists experienced a consolidation. Network restructuring may demonstrate an adaptive response that ensures patient continuity of care, but it may also reflect unique challenges faced by oncologists practicing in rural versus urban settings.Supplementary InformationThe online version contains supplementary material available at 10.1007/s41109-025-00762-3.

Some types of Internet security countermeasures require cooperative adoption to effectively defend against cyberattacks, such as authentication mechanisms that prevent spoofing attacks. In these mechanisms, both the information source and the receiver must participate: the source signs the information, and the receiver verifies it. Adoption by only one party is insufficient, creating a social dilemma where cooperative adoption maximizes social welfare, but individual incentives diminish if others do not adopt. Resource Public Key Infrastructure (RPKI) is an example of such a mechanism, protecting against spoofing in Internet routing. Its effectiveness depends on the adoption rate across network paths and has been hindered by a low adoption ratio. However, recent trends show a moderate increase in RPKI adoption. This paper investigates the reasons behind this adoption, considering the roles of altruism, bounded rationality, and the influence of network externality. Through multi-agent simulations, we explore how these factors interact. Our findings suggest that while bounded rationality is crucial for overcoming the social dilemma, the effect of altruism is limited, and the underlying network externality also plays a significant role in facilitating spontaneous adoption across the network. These insights have implications for both network operators and policymakers in designing strategies to promote the widespread adoption of security technologies.

People’s social networks are fragmenting and partisan animosity is rising. Three studies highlight the connection between social network density and partisan animosity. Americans with denser networks (where more social contacts know each other) appear to often harbor less animosity toward political opponents. One explanation for this association is feelings of unity: dense social networks feel more unified, and people project these feelings of connection and common identity onto distant fellow Americans. Study 1 supports this idea with a representative sample of Americans: social network density predicts feelings of local unity, which predicts feelings of national unity, which predicts warmth toward and willingness to interact with political opponents. Study 2 attempts to manipulate perceived social network density and finds a “causal cascade” of indirect effects culminating in lower partisan animosity, suggesting that more involved interventions may reduce partisan animosity. Study 3 uses nationwide county-level data from the Social Capital Atlas and Project Implicit to reveal that communities with higher social network density exhibit lower partisan animosity. The results suggest that considering the overarching structures of social networks is crucial to understanding partisan animosity, and that fostering social cohesion could help mend political divides.

Nestedness is a property of bipartite complex networks that has been shown to characterize the peculiar structure of biological and economical networks. Emergence of nestedness is commonly due to two different schemes: (i) mutualistic behavior of nodes, where nodes of each class have an advantage in associating with each other, such as plant pollination or seed dispersal networks; (ii) geographic distribution of species, captured in a so-called biogeographic network where species represent one class and geographical areas the other one. Motivated by analogies with biological networks, we study the nestedness property of the public Internet peering ecosystem, an important part of the Internet where autonomous systems (ASes) exchange traffic at Internet eXchange Points (IXPs). We propose two representations of this ecosystem using a bipartite graph derived from PeeringDB data. We statistically confirm the nestedness property of both graphs. From this unique observation, we show that we can use node metrics to extract new key ASes, study IXPs’ attractiveness and make efficient prediction of newly created links over a two-year period.

Abstract The manipulation of the Impact Factor (IF) through editorial decisions inflating self-citations is a growing concern in academic publishing. This work introduces an agent-based model simulating journals as rational agents competing for IF ranking positions in a zero-sum game, generating synthetic citation networks that reproduce key patterns such as the Matthew Effect and the specialization of manipulation strategies across interconnected disciplines. By comparing IF calculation policies with and without self-citations, statistical analysis of simulation results demonstrates that excluding self-citations significantly reduces incentives for manipulative tactics, offering computational evidence for developing policies that promote scientific integrity within the academic publishing ecosystem.