Network Analysis Tools Research Articles

A complex network analysis of groundwater wells in and around the Doñana Natural Space, Spain

This paper proposes the use of Complex Network Analysis tools to study the spatial distribution and temporal evolution of groundwater abstraction wells within a region of interest. To that end a weighted undirected network connecting wells within a certain distance was built. The edge weights take into account the distance between the wells as well as their corresponding abstraction volumes. In addition to the conventional complex network metrics, two specific abstraction-based centrality indexes are proposed, based on the total impact of a groundwater abstraction well on its surrounding wells and on the total impact of surrounding wells on a specific location. The proposed approach has been applied to the area of the Doñana Natural Space, one of the most important wetlands in Europe. The topology of the network reveals a large number of small, disconnected components and a few large components. This is reflected also in the degree distribution, which follows a Power Law distribution. The network has a large average clustering coefficient, a small average path length and a low level of centralisation. Global and average local network efficiency are also high. The network is assortative in terms of degree-degree and strength-strength correlations. Eigenvector centrality of the nodes reveals the location of influential wells.

Resilience patterns in the analysis of social-ecological networks of small-scale fishing in a subtropical estuarine system

Network analysis concepts and tools are being increasingly applied to Social-ecological systems (SES) in human-environment and sustainability sciences. In this study, we used the Gao metric (βeff) to measure the resilience of networks in small-scale fisheries (SSF) in the Paranaguá Estuarine Complex (PEC). The purpose was to evaluate the variations and trends in network resilience across eight hydrological sectors of the PEC during different seasons. Our analysis revealed that the βeff metric varied significantly across the eight SES sectors, suggesting that resilience is not uniformly distributed. Notably, the sectors located internally the estuary exhibited lower resilience values than the others. The average trophic level of the fishery catches was the higher hierarchical position variable that most affected the network topological metrics, followed by variables magnitude of conflicts, water and sediment quality, and rainfall. Our hydrodynamic sectorization, reflecting ecological and social interactions, enabled us to identify distinct interaction networks within the PEC. This shows the complexity of managing these SES to ensure the sustainability and resilience of the system. While other variables and spatial and temporal scales could lead to different estimates and relationships, our findings shed light on the importance of resilience quantification and the interaction networks within SES to manage them effectively.

Mapping epistemic pluralism: A network analysis of discursive practices in communities promoting refused knowledge about healthcare and wellbeing

This article presents an analysis of discourses performed in communities that share and disseminate knowledge refused by institutional science. The study focuses on an online community concerned with alkaline water, food, and lifestyle, aiming at understanding how promoters of refused knowledge in this community enrol other forms of knowledge, including science. Theoretically, this work is framed in Science & Technology Studies, and in the tradition of actor-network theory, situating itself in a recent turn that takes epistemic instability and pluralism into due consideration, thus overcoming opaque views of the opposition between science and non-science. Empirically, this fine-grained analytic purpose is addressed by a mixed-method strategy in which discursive practices are observed through a web-ethnography conducted between January 2020 and December 2021 on the relevant online spaces and then analysed qualitatively and quantitively by means of formal techniques. Relying on the tools of social network analysis, the discursive space of the community under study is formalised as a two-mode network of knowledge claims and heterogeneous actors enrolled in discourse to sustain those claims. Then, community detection is performed to map the different assemblages of claims and actors and the relevant repertoires characterising those assemblages. Finally, multiple correspondence analysis applied to two-mode networks is used to highlight the dimensions of concern and meaning expressed in the knowledge organisation of this community.

POI VizNet: New QGIS Tool to construct Visibility Networks in Cities

Abstract. Network science and graph theory offer attractive models to describe urban phenomena and examine cases of urban planning and design. However, none of the existing network analysis tools consider simultaneously human visual perception and locations of urban activities. In this paper we introduce POI VizNet, a new QGIS plug-in that constructs various 2-dimentianal undirected graphs of unobstructed lines of sight. The plug-in integrates an increasing amount of available GIS-based data of Point Of Interest (POI) and visibility into one readily accessible analytical framework. Graphs are created between two types of decision points during urban travel – street intersections and POIs, origins, and destinations of travel, by connecting these potential observer’s decision locations in an open space between buildings. In addition, POI VizNet provides advanced options to build graphs using a predefined viewing distance and perceptual perspective. Visibility graphs are constructed and visualised as new layers in QGIS and delivered as network files suitable for further exploration, analysis, and visualisation in various network software packages.

Discursive Practices during an Electoral Cycle: Public Opinion and Political Disillusionment on Twitter

Electoral cycles are highly charged, politically intense moments that influence public discourse. Political elites, citizens, and the traditional news media seek to generate opinions and interactions on social networks. This research is motivated by the following questions: What are the communicative and deliberative practices used in these spaces? Is it possible to identify the characteristics that –through deliberative conversations– potentially foster or undermine democratic debate, particularly when using populist and polarizing discourse? Using a mixed methods approach, we apply a social network analysis tool to track conversations and identify the volume of political discourse (N=346,000). Using selective and staged filtering, we identified posts from nine Peruvian Twitter accounts during the September 2022 electoral campaign in Lima: three candidates, three media outlets, and three accounts with high levels of engagement. Our data comprised the comments from these nine accounts and was extracted using an Application Programming Interface (API). Subsequently, we carried out ethnographic content analysis on publications with more than 30 comments. In this phase, we analyzed Twitter comments using a codebook to identify deliberative practices and user responses. Our findings underscore the significant role of principal media outlets in shaping political conversation on Twitter. We also discovered that attempts to interact and deliberate were often overshadowed by heated comments attempting to impose opinions on others. Most importantly, our research reveals a widespread disillusionment with politics, political institutions, and even the political preferences of fellow citizens in Peru, highlighting a key challenge for democratic discourse on social media.

Generative AI-based knowledge graphs for the illustration and development of mHealth self-management content.

Digital therapeutics (DTx) in the form of mobile health (mHealth) self-management programs have demonstrated effectiveness in reducing disease activity across various diseases, including fibromyalgia and arthritis. However, the content of online self-management programs varies widely, making them difficult to compare. This study aims to employ generative artificial intelligence (AI)-based knowledge graphs and network analysis to categorize and structure mHealth content at the example of a fibromyalgia self-management program. A multimodal mHealth online self-management program targeting fibromyalgia and post-viral fibromyalgia-like syndromes was developed. In addition to general content, the program was customized to address specific features and digital personas identified through hierarchical agglomerative clustering applied to a cohort of 202 patients with chronic musculoskeletal pain syndromes undergoing multimodal assessment. Text files consisting of 22,150 words divided into 24 modules were used as the input data. Two generative AI web applications, ChatGPT-4 (OpenAI) and Infranodus (Nodus Labs), were used to create knowledge graphs and perform text network analysis, including 3D visualization. A sentiment analysis of 129 patient feedback entries was performed. The ChatGPT-generated knowledge graph model provided a simple visual overview with five primary edges: "Mental health challenges", "Stress and its impact", "Immune system function", "Long COVID and fibromyalgia" and "Pain management and therapeutic approaches". The 3D visualization provided a more complex knowledge graph, with the term "pain" appearing as the central edge, closely connecting with "sleep", "body", and "stress". Topical cluster analysis identified categories such as "chronic pain management", "sleep hygiene", "immune system function", "cognitive therapy", "healthy eating", "emotional development", "fibromyalgia causes", and "deep relaxation". Gap analysis highlighted missing links, such as between "negative behavior" and "systemic inflammation". Retro-engineering of the self-management program showed significant conceptual similarities between the knowledge graph and the original text analysis. Sentiment analysis of free text patient comments revealed that most relevant topics were addressed by the online program, with the exception of social contacts. Generative AI tools for text network analysis can effectively structure and illustrate DTx content. Knowledge graphs are valuable for increasing the transparency of self-management programs, developing new conceptual frameworks, and incorporating feedback loops.

Weak dangling block reordering and multi-step block compression for efficiently computing and updating PageRank solutions

The PageRank model is a powerful tool for network analysis, utilized across various disciplines such as web information retrieval, bioinformatics, community detection, and graph neural network. Computing this model requires solving a large-dimensional linear system or eigenvector problem due to the ever-increasing scale of networks. Conventional preconditioners and iterative methods for general linear systems or eigenvector problems often exhibit unsatisfactory performance for such problems, particularly as the damping factor parameter approaches 1, necessitating the development of specialized methods that exploit the specific properties of the PageRank coefficient matrix. Additionally, in practical applications, the optimal settings of the hyperparameters are generally unknown in advance, and networks often evolve over time. Consequently, recomputation of the problem is necessary following minor modifications. In this scenario, highly efficient preconditioners that significantly accelerate the iterative solution at a low memory cost are desirable. In this paper, we present two techniques that leverage the sparsity structures and numerical properties of the PageRank system, as well as a preconditioner based on the computed matrix structure. Experiments demonstrate the positive performance of the proposed methods on realistic PageRank computations.

Systems Modeling of the Water-Energy-Food-Ecosystems Nexus: Insights from a Region Facing Structural Water Scarcity in Southern Spain

The complex relationship between water, energy, food, and ecological systems, known as the WEFE nexus, has emerged as a major topic in the debate about sustainable economic development and resource management. This subject is of special interest in Mediterranean coastal areas as rapid economic expansion driven by population growth, higher influx of tourists, and intensification of agriculture is leading to structural water scarcity conditions. However, addressing the diverse range of issues associated with the nexus is a difficult task due to the existence of intricate interconnections, interdependencies, and nonlinearities within and across its various components. Accordingly, this case study applies a combination of participatory systems modeling and network analysis tools to yield insights into the complexity of this nexus in Axarquia, a region with features that make it an example of water-stressed jurisdictions in the Mediterranean. Overall, our results provide a strong foundation to understand the dynamics that govern this nexus in regions where the availability of freshwater resources is a significant concern. Furthermore, they lay the groundwork for the development of models and scenarios to simulate the impact of various policies and interventions on the overall system.

From CFOs to crypto: exploratory study unraveling factors in corporate adoption

Cryptocurrency adoption has gained significant attention across various fields owing to its disruptive potential and associated challenges. However, companies' adoption of cryptocurrencies remains relatively low. This study aims to comprehensively examine the factors influencing cryptocurrency adoption, their interrelationships, and their relative importance. To achieve this objective, we employ a Decision-Making Trial and Evaluation Laboratory (DEMATEL) approach coupled with network analysis tools. By adopting a practical approach rather than a purely theoretical one, our unique contribution lies in the valuable insights derived from experienced Chief Financial Officers (CFOs) of various companies with experience in both traditional finance and cryptocurrencies. Furthermore, the unique blend of analytical rigor and industry expertise supports the study's relevance, offering nuanced insights that are not only academically robust but also immediately applicable in the corporate landscape. Our findings highlight the paramount importance of safety in transactions and trust in the chosen platform for companies considering cryptocurrency adoption. Additionally, criteria such as faster transactions without geographical limitations, lower transaction fees, seamless integration with existing systems, and potential cost savings are identified as crucial drivers. Both the DEMATEL approach and network analysis reveal strong interconnections among the criteria, emphasizing their interdependence and, notably, their reliance on transactional safety. Furthermore, our causes and effects analysis indicates that CFOs perceive company-led cryptocurrency adoption to positively impact the broader cryptocurrency market.

Fitness-based growth of directed networks with hierarchy

Abstract Growing attention has been brought to the fact that many real directed networks exhibit hierarchy and directionality as measured through techniques like Trophic Analysis and non-normality. We propose a simple growing network model where the probability of connecting to a node is defined by a preferential attachment mechanism based on degree and the difference in fitness between nodes. 
In particular, we show how mechanisms such as degree-based preferential attachment and node fitness interactions can lead to the emergence of the spectrum of hierarchy and directionality observed in real networks. In this work, we study various features of this model relating to network hierarchy, as measured by Trophic Analysis. This includes (I) how preferential attachment can lead to network hierarchy, (II) how scale-free degree distributions and network hierarchy can coexist, (III) the correlation between node fitness and trophic level, (IV) how the fitness parameters can predict trophic incoherence and how the trophic level difference distribution compares to the fitness difference distribution, (V) the relationship between trophic level and degree imbalance and the unique role of nodes at the ends of the fitness hierarchy and (VI) how fitness interactions and degree-based preferential attachment can interplay to generate networks of varying coherence and degree distribution. We also provide an example of the intuition this work enables in the analysis of a real historical network. This work provides insight into simple mechanisms which can give rise to hierarchy in directed networks and quantifies the usefulness and limitations of using Trophic Analysis as an analysis tool for real networks.

Identifying authorial roles in research: A Kano model-based bibliometric analysis for the Journal of Medicine (Baltimore) 2023.

The landscape of research roles within academic journals often remains uncharted territory, with authorial contributions frequently reduced to linear hierarchies (e.g., professor and assistant professor). The Kano model, traditionally used in customer satisfaction research, offers a nuanced framework for identifying the multifaceted roles of authors in scholarly publications. This study utilizes the Kano model to dissect and categorize the roles of authors in the medicine field. To conform to the hypothesis, China is the research leader while the US is the research collaborator, as reflected in the publications of the journal of Medicine (Baltimore) in the year 2023. We conducted a comprehensive bibliometric analysis of all research articles published in the journal of Medicine (Baltimore) in 2023. The Kano model was applied to classify authors into 5 categories reflective of their research roles: followers, leaders, partners, contributors, and collaborators. Data on author publications and co-authorship networks with multi-author rates (MARs) were analyzed to assign Kano categories based on the authorship positions of first and corresponding authors. Descriptive statistics and network analysis tools were used to interpret the data, including radar plots, geographical maps, and Kano diagrams. The analysis covered 1976 articles, uncovering a complex network of author roles that extends beyond the conventional binary distinction of lead and supporting authors (i.e., leading, and following researchers). A research leader in China and a collaborator in the US were conformed to support the hypothesis, based on their publications (1148 vs 51) and MARs (12.20% vs 19.61%). The Kano classification was visually adapted to classify authors (or entities) into 5 categories. The combined choropleth and geographical network maps were illustrated to identify author roles in research briefly. The Kano model serves as an effective tool for uncovering the diverse contributions of authors in medical research. By moving beyond the lead and follower dichotomy, this study highlights the intricate ecosystem of authorial roles, emphasizing the importance of each in advancing knowledge within the field of medicine. Future application of the Kano model could foster a more collaborative and inclusive recognition of contributions across various disciplines.

Ensemble of artificial intelligence and physically based models for rainfall–runoff modeling in the upper Blue Nile Basin

ABSTRACT This study investigated the performance of the adaptive neuro-fuzzy inference system (ANFIS), feed forward neural network (FFNN), Soil and Water Analysis Tool (SWAT), Hydrologic Engineering Center's Hydraulic Modeling System (HEC-HMS), Hydrologiska Byråns Vattenbalansavdelning (HBV), and support vector regression (SVR) models for rainfall–runoff modeling using gauged and satellite rainfall, and their fusions in the Gilgel Abay watershed, Ethiopia. Afterward, simple average ensemble (SAE), weighted average ensemble (WAE), and neural network ensemble (NNE) techniques were applied to combine the outputs of individual models under three scenarios. The performance of the models was evaluated using Nash–Sutcliffe efficiency (NSE) and root mean square error (RMSE). The results demonstrated that the ANFIS model outperformed all the other single models with validation stage NSE values of 0.864 and 0.875, and RMSE values of 23.58 and 21.84 m3/s for gauge and fusion rainfall data, respectively. Among the physical-based models, SWAT gave better modeling performance with the validation stage NSE values of 0.81 and 0.821 for gauge and fusion rainfall data, respectively. Moreover, an ensemble of artificial intelligence and physical-based models greatly improved the overall modeling performance. The NNE improved the performance of single models up to 15.7 and 21.2 5% for fusion and satellite-based rainfall modeling, respectively.

Analyzing network flows of used EEE and e‐waste with platform data: Adding reuse into the EPR system for WEEE recycling in China

AbstractChina has established a nationwide formal e‐waste disassembly system based on the extended producer responsibility principle. However, the system mainly focuses on material recycling while excluding the reuse of used electrical and electronic equipment (used EEE) due to the complexity of flows and transactions during the reuse stage. Recently, emerging online platforms have played an increasingly crucial role in the value chain of reuse and recycling, significantly improving the visibility of flows in these activities. This paper aims to depict the spatial structure of the used EEE and e‐waste flow networks using a multi‐scale analysis framework. Using spatial analysis tools in complex network analysis, we characterize cross‐city reuse flows in China. A clear regional pattern of a hierarchical reuse network of notable regional hubs in certain regions in China is revealed in this analysis. The role of regulation is demonstrated in the comparison of the spatial flows in reuse and recycling. In conclusion, it is proposed that reuse should be included and emphasized in the management of waste electrical and electronic equipment in China to maximize the value of circularity.

Empowering Equality Law With Nlp: A Social Network Analysis Approach to Drive Legal SDGS

Objectives: The purpose of this study is to use social media analysis to understand people's opinions about legal equality. Theoretical Framework: The study builds on social studies theory and existing literature to explore the complexity of society's reaction to legal equality. It focuses on how public opinion evolves over time and the factors influencing these changes, including language, age, and gender. Method: Data was gathered from various social media platforms on September 8, 14, and 15, 2023, using the Talkwalker Social Intelligence Platform. The analysis involved monitoring public opinion trends and performing a theme analysis to identify recurring themes in the discourse. Results and Discussion: The findings revealed essential trends in societal reactions to legal equality. Notable changes in perceptions of specific issues were observed, and prominent aspects of legal equality, such as language, age, and gender, were identified. These results provide a comprehensive understanding of how society perceives legal equality and the dynamics influencing these perceptions. Research Implications: The study enhances existing literature and advances social studies theory by offering insights into the complexity of societal reactions to legal equality. It also demonstrates the potential of social media analysis using Social Network Analysis (SNA) and Talkwalker-type tools to understand how social interactions shape opinions and to inform policy planning. Originality/Value: This research illuminates the social dynamics of legal equality and showcases the effectiveness of combining SNA with social media analysis tools. The findings offer valuable contributions to theoretical and practical discussions on legal equality and policy-making, particularly in the context of promoting the SDGs.

A multilevel backbone extraction framework

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.

Quantifying Women’s Marginalisation in Ibero-American Film Culture During the First Half of the Twentieth Century: A Network-Science Proposal

The research presented here uses the tools of social network analysis to empirically show a socio-cultural phenomenon already addressed by the social sciences and history: the historical marginalisation of women in the field of cinema. The novelty of our approach lies in the use of a large amount of heterogeneous historical data. On the one hand, we built a network of interactions between people involved in the film field in Ibero-America during the first half of the twentieth century. On the other hand, we propose a k-core decomposition and a multi-layered analysis, as a quantitative way to study the position of women within the cultural melieu. After conducting our analysis, we concluded that women were mostly situated in the outer k-shells of the empirical network, and their distribution was not uniform across the k-shells. From a qualitative perspective, these results can be interpreted as the consequence of the lack of evidence of the participation of women in the public sphere.

EVALUATION OF HEALTH FACILITIES LOCATION TO SUPPORT ACCESSIBILITY TO ESSENTIAL HEALTH SERVICES IN BANDAR LAMPUNG CITY, LAMPUNG PROVINCE, INDONESIA

Health remains one of the most important aspects of every human life, with access to quality healthcare services being a pivotal component of human well-being. In Indonesia, health facility distribution has been characterized by disparities, primarily concentrated on the populous Java Island. This distribution inequity underscores a significant challenge in achieving Sustainable Development Goals (SDGs) under Goal 3, which aims to ensure healthy lives and promote well-being for all. Bandar Lampung, the most densely populated city in Lampung Province and one of Sumatra's most populated cities, experiences an intensified demand for healthcare facilities due to its substantial population. However, the city grapples with numerous infrastructural challenges, including suboptimal road conditions that hinder accessibility to essential healthcare services. These challenges further underscore the urgency of assessing and addressing the healthcare infrastructure within Bandar Lampung. This research seeks to undertake a comprehensive evaluation of health facility availability in Bandar Lampung City, as well as an in-depth analysis of the accessibility of healthcare services in different areas of the city. To achieve this, the study employs the Network Analysis Tool, which leverages road network data to model travel distances and times. In addition to road network data, geospatial population data sourced is integrated into the analysis to gauge the demand for healthcare facilities within Bandar Lampung. This multifaceted approach is essential in comprehensively addressing the complex interplay between supply and demand in the healthcare sector. The combination of these datasets culminates in the creation of an accessibility model that unveils the disparities in healthcare access across Bandar Lampung. The model highlights areas where accessibility to healthcare facilities remains surprisingly low, despite the city's high population density. Conversely, it also identifies regions with high accessibility, potentially serving as templates for improving healthcare distribution in underserved areas. By revealing these disparities, this research contributes valuable insights to healthcare planning and policy formulation in Bandar Lampung and offers a template for addressing similar challenges in other regions with uneven healthcare distribution.

Influential simplices mining via simplicial convolutional networks

The identification of influential simplices is crucial for understanding higher-order network dynamics. Yet, despite relatively mature research on influential nodes (0-simplices) mining, characterizing simplices’ influence and identifying influential simplices remain challenging due to notable discrepancies in vital nodes and vital simplices mining. In this paper, we propose a higher-order graph learning model, named influential simplices mining neural networks (ISMnet), to identify vital simplices in simplicial complexes. ISMnet leverages novel higher-order representations: hierarchical bipartite graphs and higher-order hierarchical (HoH) Laplacians, where target simplices are grouped into a hub set and can interact with other simplices. It also employs learnable graph convolution operators in each HoH Laplacian domain to capture interactions among simplices and can identify influential simplices of arbitrary order by changing the hub set. Notably, ISMnet addresses the limitations inherent in traditional graph neural networks that struggle with higher-order tasks, while seamlessly retaining the capability to exploit network topology and node features concurrently. Numerical results on empirical and synthetic datasets demonstrate that ISMnet significantly outperforms existing methods by at least 12% and 4%, respectively, in ranking 2-simplices. In general, this novel framework promises to serve as a potent tool in higher-order network analysis.

Modeling pedestrian activity in cities with urban network analysis

The global climate-change crisis, along with public health and economic competitiveness challenges in cities around the world have underscored the need for analytic tools to examine the relationship between city design and sustainable mobility. Car-centered travel demand models and land-use-transportation interaction models have historically analyzed zone-to-zone trips along major roadways, largely omitting pedestrian and bicycle trips and creating a gap in the ability for planners and urban designers to systematically assess non-motorized outcomes of development interventions. We present the Urban Network Analysis tools to address this gap. UNA tools offer an accessibility-based framework for analyzing how built environments influence pedestrian travel in both existing and newly planned built environments. Developed as a free plugin for Rhinoceros 3D since 2015 and applied in several cities and research projects internationally, this paper describes the current Urban Network Analysis modeling framework and discusses the unique contributions the framework offers compared to existing pedestrian modeling approaches. Using Somerville, MA as case example, we demonstrate several commonly used functions for planners: examining pedestrian accessibility over networks; identifying critical walking routes to destinations; estimating foot-traffic on street segments; identifying frustration points for pedestrians; and evaluating how development changes may impact pedestrian activity in their vicinity. Such analyses can provide analytic evidence to pedestrian infrastructure planning and investment, and enable planners, designers, and policy makers to prioritize projects that increase sustainable mobility outcomes.

NetDiffusion: Network Data Augmentation Through Protocol-Constrained Traffic Generation

Datasets of labeled network traces are essential for a multitude of machine learning (ML) tasks in networking, yet their availability is hindered by privacy and maintenance concerns, such as data staleness. To overcome this limitation, synthetic network traces can often augment existing datasets. Unfortunately, current synthetic trace generation methods, which typically produce only aggregated flow statistics or a few selected packet attributes, do not always suffice, especially when model training relies on having features that are only available from packet traces. This shortfall manifests in both insufficient statistical resemblance to real traces and suboptimal performance on ML tasks when employed for data augmentation. In this paper, we apply diffusion models to generate high-resolution synthetic network traffic traces. We present NetDiffusion, a tool that uses a finely-tuned, controlled variant of a Stable Diffusion model to generate synthetic network traffic that is high fidelity and conforms to protocol specifications. Our evaluation demonstrates that packet captures generated from NetDiffusion can achieve higher statistical similarity to real data and improved ML model performance than current state-of-the-art approaches (e.g., GAN-based approaches). Furthermore, our synthetic traces are compatible with common network analysis tools and support a myriad of network tasks, suggesting that NetDiffusion can serve a broader spectrum of network analysis and testing tasks, extending beyond ML-centric applications.