Embedded Files
Goals
Goals
Design and develop a parallel, scalable, extendable, software platform, CANDY, for updating properties of dynamic networks.
Provide functionality and tools to create new algorithms or modify existing ones, catering to users with varying expertise.
Implement CANDY platform on distributed and shared memory systems, and GPUs, providing user-friendly interfaces.
Challenges
Challenges
Efficiently analyzing dynamic networks for real-life applications.
Comprehensive cyberinfrastructure to support innovative research in large dynamic networks.
Creating high-performance software with performance optimization.
Intellectual Merit
Intellectual Merit
Create a novel hierarchical taxonomy of network analysis algorithms that allows for layered specification of parallel algorithms based on multiple parameters.
Develop templates for creating new scalable, parallel algorithms for dynamic network analysis.
Design algorithms to partition streaming set of nodes and edges into network snapshots at changing points.
Propose invariant-based quantifiable performance metrics for analyzing large dynamic networks.
Evaluate CANDY software on genomic data, and cost-effective operation of complex mining applications.
Broader Impacts
Broader Impacts
Collaborate with multidisciplinary national/international research groups to evaluate the effectiveness of the developed platform, algorithms and software tools.
Host workshops, webinars, and tutorials to educate research and broader community about CANDY.
Disseminate project outcomes via a dedicated website, invited talks, demos, and high-quality publications.
Train next generation data scientists to develop CANDY platform; engage underrepresented minority, women, and K-12 students.
Page updated
Google Sites
Report abuse