User Tools

Site Tools


projects:year3:14.3

14.3 - Visual Analytic Approaches for Mining Large-Scale Dynamic Graphs

Project - Summary

Objectives:

The primary goal of the project is to develop and demonstrate a visual analytics framework for large-scale time-evolving graphs. We designed and implemented a prototype system, we investigated integration-related issues pertaining to graph analysis, visualization, and interactive touch interfaces, and we developed techniques to address integration problems and to improve the overall performance of visual analytics system.

Project - Team

Team Member Role Email Phone Number Academic Site/IAB
Raju Gottumukkala PI Not available Not available UL Lafayette
Chrisoph Borst Co-PI Not available Not available UL Lafayette
Siva Venna Graduate Student Not available Not available UL Lafayette
Nicholas Lipari Graduate Student Not available Not available UL Lafayette

Project - Impact and Uses/Benefits

Several companies in the business intelligence domain such as SAS, IBM, SPSS are pursuing novel ways to improve their data presentation through new products such as SAS Visual Analytics, and IBM's Many Eyes. Moreover, several new BI tools such as Tableau, Birst and Google Fusion Tables are also providing some basic interactive visualization capabilities. While these tools provide some basic visualization and interaction capabilities for users to interact with the data, these tools are far from promoting analytics discourse with the visualization environment.

Real-world phenomenon - such as the evolution of social community networks, infrastructure networks, epidemiology networks, and IP traffic networks can be modeled as time-evolving graphs using data from real-time data sources. The time-dimension of these graphs introduces additional complexity to process, store, analyze, and visualize the data - which cannot be handled efficiently by existing decision support tools.Most of the existing tools summarize or provide aggregate statistics on these time evolving graphs. Hence it is very labor intensive and time consuming to navigate through historical data to see emerging patterns. The visual analytics framework and prototype we provided provides new knowledge in terms of designing visual analytics platforms for handling high volume, high velocity, and highly relational data that has board applications for investigative analysis – especially in cybersecurity, infrastructure surveillance, homeland security, and emergency management.

Project - Deep Dive

Project - Documents

FilenameFilesizeLast modified
14.3_year_3_executive_summaries_combined.pdf441.9 KiB2019/08/22 11:26
14.3_year_3_cvdi_ip_letter_combined.pdf769.4 KiB2019/08/22 11:26
14.3_year_3_presentation.pptx1.6 MiB2019/08/22 11:26
14.3_year_3_final_report.pdf2.0 MiB2019/08/22 10:21
projects/year3/14.3.txt · Last modified: 2019/08/22 10:22 by sally.johnson