Collaborative Research: HNDS-I: IDEANet: Integrating Data Exchange and Analysis of Networks

点击翻译按钮获取中文摘要

In this project, the Integrated Data Exchange and Analysis for Networks (IDEANet) platform is constructed with the aim of solving several problems that have confronted researchers using network analysis. Over the last 30 years, network analysis — a method for measuring and modeling the patterns of connections between objects (people, places, proteins, etc.) — has emerged as a prominent data-intensive research theme with broad applications and with significant Federal research investment. Network analysis is unique amongst methodologies in its focus on interdependence, allowing investigators to understand how, for example, diseases spread through populations, hostile groups organize, or predator/prey patterns of species shape an ecosystem, as well as myriad other substantive applications. Despite the wide use of network analysis, the analytical and computational tools are highly specialized and often have domain-specific idiosyncrasies that severely limit integration across fields and creates unnaturally high barriers to entry for investigators new to network analysis who want to apply these methods to their problems. Moreover, current computational tools manage data in ways that make errors undetectable and easy to make, lowering rigor and reproducibility in network analysis studies. The new computational toolkit and data storage framework developed in IDEANet aims to solve these problems through its integration of analytic methods and data archiving, discovery, and distribution. To make network data easier to find, use and share, four integrated objectives are pursued in IDEANet: (1) Provision of an easy-to-use software tool, integrated via the R statistical programming language, with methodologically state-of-the-art analysis routines and metrics, developed in parallel with a ShinyR app that allows GUI use for R novices; (2) Construction of a network data translation engine that allows users to easily move data between the myriad extant formats found across application fields; (3) Provision of a core metrics compute engine that automatically generates vetted best practice metrics on network data of multiple types; (4) Development of a data repository and archival system that leverages the compute tools to build harmonized network data releases in multiple formats with pre-computed summary statistics and metrics, tailored to different levels of security. The secure-data archive capacity makes use of ImPACT (Infrastructure for Privacy-Assured CompuTations) distributed secure data services developed under prior NSF investment (#1659367). This award by the Directorate for Social, Behavioral, and Economic Sciences is jointly supported by the NSF Office of Advanced Cyberinfrastructure.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.