1 17 Visualization of GTD and Multimedia Remco

1 / 17 Visualization of GTD and Multimedia Remco Chang Charlotte Visualization Center UNC Charlotte

2 / 17 Visual GTD Flow Chart Dimensional Relationships (Parallel. Sets) Entity Relationships (Geo-temporal Vis) Entity Analysis (Search By Example)

3 / 17 WHO – Terrorist Groups What Five Flexible Entry Components WHERE~ WHEN

4 / 17 Seeing Patterns… FARC showing an outlier Unusual temporal pattern of NPA

5 / 17 Parallel Sets View • Parallel Sets – Displays relationships among categorical dimensions – Shows intersections and distributions of categories

6 / 17 Parallel Sets View • Dynamic filtering on continuous dimensions can show more information • Here we see the large proportion of facility attacks and bombings in Latin America during the early 1980 s

7 / 17 Parallel. Sets - Framing

8 / 17 Entity Comparison • Uses the algorithm “Longest Common Subsequence” (LCS) to identify similar patterns

9 / 17 Grouping using MDS in 2 D • Each o represents a terrorist group • Groups form cluster according to naturally occurring trend sizes • Clusters are easily visible MDS Analysis by Country

10 / 17 Auto Video Extraction

11 / 17 Multimedia Visual Analysis

12 / 17 Concept Graph

13 / 17 Video Analysis Example CNN Fox News MSNBC • News contains view points and opinions • Find local, regional, national, and international reports of the same event to get a complete picture

14 / 17 News Lens

Integrating Terrorism Data Analysis and News Analysis 15 / 17 Terrorism Visual Analysis Terrorism Databases Terrorism VA Stab/ TIBOR Reasoning Environment Jigsaw NVAC Framing, Broadcast Affective Analysis News VA Visual Analysis News Story Databases

16 / 17 Future Work • Event-based video analysis • Smart Visual GTD – Collaboration with Daniel Kiem (Univ Konstanz, Germany) – Multimedia Analysis • Collaboration with PNNL (A. Sanfilipo, W. Pike) • Analyzes (layout of) webpages, videos, images, and unstructured texts. • Tracking temporal changes

17 / 17 Questions? Thank you! rchang@uncc. edu http: //viscenter. uncc. edu

18 / 17 Backup

19 / 17 Entity Comparison • Two strings of data (each representing a series of events) – GATCCAGT – GTACACTGAG • Basic algorithm returns length of longest common subsequence: 6 • Can return trace of subsequence if desired: – GTCCAG • GATCCAGT • GTACACTGAG • Additional variations can take into account event gap penalties, time gap penalties, and exploration of shorter, or alternate, common subsequences

20 / 17 Parallel. Sets - Framing