Many applications across domains involve the generation or analysis of networks with a very large number of interacting components. Examples include mining social networks (e.g., Twitter) or simulating them to predict spreading events (e.g., of pathogens in an epidemic or of information in a word-of-mouth network); understanding the core patterns and dynamics of biological networks (e.g., regulatory networks, protein-protein interaction); or analyzing and improving the functioning of cities (e.g., transportation networks). The size of the networks is a major challenge through all these applications. Algorithms or implementations that worked on small datasets may not be able to cope with large networks. For instance, it becomes challenging to measure the centrality of nodes or identify communities using conventional algorithms. Similarly, simulating the social interactions of extremely large populations would exceed the capacity of a small computer and thus requires venturing into distributed and parallel algorithms that leverage clusters or cloud-based resources.
To provide efficient scalable solutions, there is a documented need to develop new algorithms (e.g., parallel and distributed network algorithms, approximation algorithms), improve their implementation (e.g. via cloud- or edge-based network simulation and analysis), and identify bottlenecks through case studies.
The aim of this Research Topics to address these challenging needs based on emerging trends at the convergence of Network Science with Big Data Analytics, Large-Scale Simulation, and Machine/Deep Learning. As the generation and analysis of networks are closely related, we are interested in insightful case studies, novel algorithms, or implementations that improve our ability to quickly generate and use large-scale networks (e.g. to simulate a diffusion, cascading behaviours, or consensus dynamics), or efficiently perform fundamental tasks in network science such as centrality, community detection, and link prediction.
We invite researchers to provide solutions and case studies focusing on large-scale networks. Potential topics include, but are not restricted to:
- Case studies that characterize scalability issues of current solutions in an application domain;
- Novel algorithms for large-scale network analytics (e.g. centrality, community detection, link prediction);
- Novel network models (e.g., hierarchical, fractal, small-world, scale-free) that can be generated efficiently;
- Large-scale network simulations (e.g., simulating social interactions of an entire population, brain simulation, Internet of Things);
- Parallel hardware and software systems for efficient storing and processing of large-scale networks during simulation or analysis;
- Network simulations or analyses utilizing massive HPC, Exascale systems, cloud- or edge-computing.
We accept manuscripts of the following types: Original Research, Systematic Review, Perspective, or Brief Research Report.
Many applications across domains involve the generation or analysis of networks with a very large number of interacting components. Examples include mining social networks (e.g., Twitter) or simulating them to predict spreading events (e.g., of pathogens in an epidemic or of information in a word-of-mouth network); understanding the core patterns and dynamics of biological networks (e.g., regulatory networks, protein-protein interaction); or analyzing and improving the functioning of cities (e.g., transportation networks). The size of the networks is a major challenge through all these applications. Algorithms or implementations that worked on small datasets may not be able to cope with large networks. For instance, it becomes challenging to measure the centrality of nodes or identify communities using conventional algorithms. Similarly, simulating the social interactions of extremely large populations would exceed the capacity of a small computer and thus requires venturing into distributed and parallel algorithms that leverage clusters or cloud-based resources.
To provide efficient scalable solutions, there is a documented need to develop new algorithms (e.g., parallel and distributed network algorithms, approximation algorithms), improve their implementation (e.g. via cloud- or edge-based network simulation and analysis), and identify bottlenecks through case studies.
The aim of this Research Topics to address these challenging needs based on emerging trends at the convergence of Network Science with Big Data Analytics, Large-Scale Simulation, and Machine/Deep Learning. As the generation and analysis of networks are closely related, we are interested in insightful case studies, novel algorithms, or implementations that improve our ability to quickly generate and use large-scale networks (e.g. to simulate a diffusion, cascading behaviours, or consensus dynamics), or efficiently perform fundamental tasks in network science such as centrality, community detection, and link prediction.
We invite researchers to provide solutions and case studies focusing on large-scale networks. Potential topics include, but are not restricted to:
- Case studies that characterize scalability issues of current solutions in an application domain;
- Novel algorithms for large-scale network analytics (e.g. centrality, community detection, link prediction);
- Novel network models (e.g., hierarchical, fractal, small-world, scale-free) that can be generated efficiently;
- Large-scale network simulations (e.g., simulating social interactions of an entire population, brain simulation, Internet of Things);
- Parallel hardware and software systems for efficient storing and processing of large-scale networks during simulation or analysis;
- Network simulations or analyses utilizing massive HPC, Exascale systems, cloud- or edge-computing.
We accept manuscripts of the following types: Original Research, Systematic Review, Perspective, or Brief Research Report.