DIA2: Web-based Cyberinfrastructure for Visual Analysis of Funding Portfolios

Madhavan, Krishna; Elmqvist, Niklas; Vorvoreanu, Mihaela; Chen, Xin; Wong, Yuet Ling; Xian, Hanjun; Dong, Zhihua; Johri, Aditya

doi:10.1109/tvcg.2014.2346747

Cited by 20 publications

(11 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In other cases metadata can be used to create visualizations without direct data access [MEV14], but metadata does not exist for our domain. While researchers are sometimes able to obtain surrogate data with similar characteristics to the real data, such as Walker et al [WSD * 13] who used online business reviews in place of human-terrain reports, there are domains where no such surrogate data exist, such as in vulnerability analysis.…”

Section: Recommendationsmentioning

confidence: 99%

A Shot at Visual Vulnerability Analysis

Kerzner

Butler

Hansen

et al. 2015

Computer Graphics Forum

View full text Add to dashboard Cite

Figure 1: We present Shotviewer, software that supports visual analysis of spatial and nonspatial ballistic simulation data for vulnerability analysis. It consists of three linked views: a) the Shotline View displays an abstract representation of shots' paths through a vehicle; b) the Geometry View shows shots' 3D spatial context; and c) the System View visualizes the propagation of damage through a vehicle's systems. In this example, the green shot damages the vehicle's radio which impacts the vehicle's mobility and firepower capabilities. AbstractIncreasing the safety of vehicles is an important goal for vehicle manufacturers. These manufacturers often turn to simulations to understand how to improve a vehicle's design as real-world safety tests are expensive and time consuming. Understanding the results of these simulations, however, is challenging due to the complexity of the data, which often includes both spatial and nonspatial data types. In this design study we collaborated with analysts who are trying to understand the vulnerability of military vehicles. From this design study we contribute a problem characterization, data abstraction, and task analysis for vehicle vulnerability analysis, as well as a validated and deployed tool called Shotviewer. Shotviewer links 3D spatial views with abstract 2D views to support a broad range of analysis needs. Furthermore, reflection on our design study process elucidates a strategy of viewdesign parallelism for creating multiview visualizations, as well as four recommendations for conducting design studies in large organizations with sensitive data.

show abstract

Section: Recommendationsmentioning

confidence: 99%

A Shot at Visual Vulnerability Analysis

Kerzner

Butler

Hansen

et al. 2015

Computer Graphics Forum

View full text Add to dashboard Cite

show abstract

“…3 Here, we focus on the system's overall perceived utility. Although usability is one necessary component of utility, it is by no means sufficient.…”

Section: Dia2mentioning

confidence: 99%

Using Visualization to Derive Insights from Research Funding Portfolios

Molnar

McKenna

Liu

et al. 2015

IEEE Comput. Grap. Appl.

Self Cite

View full text Add to dashboard Cite

O ver the past few years there has been increased interest in open data access. However, data retrieval represents a major bottleneck for many organizations, including funding agencies such as the US National Science Foundation (NSF). This is due to the multimodal, high-volume nature of funding data that is often disconnected and diffuse. In effect, deriving meaningful insights from an existing funding portfolio is challenging. Issues arise not only in data representation, but also in the way that data is accessed and communicated to the end user. Providing multiple ways to visualize data, enabling users to "see" relationships, gaps, or other connections, could lead to new insights that facilitate a better overview of existing funded research and more informed decision making for future funding.As with any federal funding agency, there is complexity of data within NSF's portfolio, and it increases continuously as new proposals get submitted, reviewed, and funded or declined. NSF awards approximately 11,000 grants a year with an average performance duration of three years. The NSF staff includes 1,400 career employees, 200 scientists from research institutions on temporary duty (rotators), and 450 contract workers and staff (data obtained from www.nsf.gov). Given the relatively high percentage of turnover among temporary NSF staff, there are always new individuals that need to be trained on funding processes. It is crucial for them to familiarize themselves quickly with the funding portfolio. For example, science assistants are typically hired for a period of two years and are expected to help program of cers with queries on the data using various criteria. Because information increases continuously, and the staff need to make decisions in a short period of time, there is a need to make the existing portfolio easy to access and understand.In this environment, however, the gap between data visualization and actual insights can be difcult to bridge. To address this problem, we developed Deep Insights Anytime, Anywhere (DIA2), a data-mining and Web-based platform that uses metadata about funding information from the NSF to produce interactive data visualizations. 1 DIA2 provides an overview in an easy-to-understand format, allowing users to search, view, and analyze the NSF funding portfolio.This article presents an assessment of DIA2's usability. We attempted to evaluate its perceived utility among the target user group. Speci cally, we inquired whether and how the system leads to insights that can facilitate further decision making, such as estimates about the impact of speci c research projects. Our results show that DIA2 has good usability. Furthermore, participants identied several indicators of impact as a result of the visualizations that can be realized through DIA2. DIA2During the design and implementation of DIA2, we followed an iterative, user-centered design process. 2 To determine DIA2 requirements, we

show abstract

“…All of these metrics are essential to better understand "why users do what they do and how they leverage multiple types and instances of CI resources" (Katz et al, 2011). In recent years, there has also been growing interest in measuring the impact of CI on scientific research and publication outcome (e.g., Madhavan et al, 2014, Knepper and Börner, 2016, Fulton et al, 2017. Such insights are particularly relevant for campus CI that requires significant investment and long-term strategic and financial planning.…”

mentioning

confidence: 99%

“…While most monitoring tools for HPC are traditionally "largely passive and local in nature" (Furlani et al, 2013a,b), these new frameworks, such as Deep Insight Anytime, Anywhere (DIA2) (Madhavan et al, 2014), and Extreme Science and Engineering Discovery Environment (XSEDE) Metrics on Demand (XDMoD VA) (Fulton et al, 2017), are open source, customizable systems with "increased functionality, an improved interface, and high-level charting and analytical tools" (Palmer et al, 2015). DIA2 is a web-based visual analytics system aiming to assess research funding portfolios (Madhavan et al, 2014). Open XD metrics on demand (XDMoD) is one of the most widely used software systems in the US and is currently employed by more than 200 institutions to evaluate their HPC usage (Palmer et al, 2015;Fulton et al, 2017).…”

mentioning

confidence: 99%

XD Metrics on Demand Value Analytics: Visualizing the Impact of Internal Information Technology Investments on External Funding, Publications, and Collaboration Networks

Scrivner

Singh

Bouchard

et al. 2018

Front. Res. Metr. Anal.

View full text Add to dashboard Cite

Many universities invest substantial resources in the design, deployment, and maintenance of campus-based cyberinfrastructure (CI). To justify the expense, it is important that university administrators and others understand and communicate the value of these internal investments in terms of scholarly impact. This paper introduces two visualizations and their usage in the Value Analytics (VA) module for Open XD metrics on demand (XDMoD), which enable analysis of external grant funding income, scholarly publications, and collaboration networks. The VA module was developed by Indiana University's (IU) Research Technologies division, Pervasive Technology Institute, and the CI for Network Science Center (CNS), in conjunction with the University at Buffalo's Center for Computational Research. It provides diverse visualizations of measures of information technology (IT) usage, external funding, and publications in support of IT strategic decision-making. This paper details the data, analysis workflows, and visual mappings used in two VA visualizations that aim to communicate the value of different IT usage in terms of NSF and NIH funding, resulting publications, and associated research collaborations. To illustrate the feasibility of measuring IT values on research, we measured its financial and academic impact from the period between 2012 and 2017 for IU. The financial return on investment (ROI) is measured in terms of IU funding, totaling $339,013,365 for 885 NIH and NSF projects associated with IT usage, and the academic ROI constitutes 968 publications associated with 83 of these NSF and NIH awards. In addition, the results show that Medical Specialties, Brain Research, and Infectious Diseases are the top three scientific disciplines ranked by the number of publications during the given time period.

show abstract

DIA2: Web-based Cyberinfrastructure for Visual Analysis of Funding Portfolios

Cited by 20 publications

References 26 publications

A Shot at Visual Vulnerability Analysis

A Shot at Visual Vulnerability Analysis

Using Visualization to Derive Insights from Research Funding Portfolios

XD Metrics on Demand Value Analytics: Visualizing the Impact of Internal Information Technology Investments on External Funding, Publications, and Collaboration Networks

Contact Info

Product

Resources

About