1. Simplifying Network Visualizations with Motif Glyphs Cody Dunne and Ben Shneiderman {cdunne, ben}@cs.umd.edu 29 th Annual Human-Computer Interaction Lab Symposium, May 22–23, 2011College Park, MD

2. Network Analysis is Hard

3. Who Uses Network Analysis

4. So what can we do?

5. Better LayoutsAlternate Visualizations Hachul & Jünger, 2006 Navlakha et al, 2008 Graph Summarization Gove et al, 2011; Henry & Fekete, 2006; Dunne et al, 2012; Freire et al, 2010; Wattenberg, 2006

6. Lostpedia articles 2: Motifs often dominate the visualization 3: Motifs members can be functionally equivalent Observations 1: There are repeating patterns in networks (motifs)

7. Our approach: Motif Simplification to reduce visualization complexity

8. Motif Simplification Fan Motif2-Parallel Motif

9. Lostpedia articles

11. Glyph Design Guidelines Representative: topology, count & attributes Easily distinguishable Easily comparable Allow overlaps

12. Glyph Design: Fan

13. Glyph Design: Parallel

14. Interactivity Fan motif: 133 leaf vertices with head vertex “Theory”

15. Voson Web Crawl

22. Implemented in NodeXL

23. User Impressions “I’m overwhelmed, … this is like one of those vision tests at the eye doctor” “Now I can see the central pages…[and] pairwise connections”

24. Discussion Motif simplification effective for Reducing complexity Understanding larger relationships However Frequent motifs may not be covered Glyph design has tradeoffs Details & algorithms in Tech Report

25. Take Away Messages Motif simplification reduces the complexity of network visualizations. Free and open source next month in NodeXL : http://nodexl.codeplex.com/ http://www.cs.umd.edu/hcil/motifs/ Thanks to: Social Media Research Foundation Connected Action Consulting Group