egocentric analysis of dynamic networks with egolines
play

Egocentric Analysis of Dynamic Networks with EgoLines Jian Zhao, - PowerPoint PPT Presentation

Egocentric Analysis of Dynamic Networks with EgoLines Jian Zhao, Michael Glueck, Fanny Chevalier, Yanhong Wu, Azam Khan Dynamic Networks Time 2 Year 1 Year 2 3 Egocentric Analysis My co-authors co-author Me My co-author 4


  1. Egocentric Analysis of Dynamic Networks with EgoLines Jian Zhao, Michael Glueck, Fanny Chevalier, Yanhong Wu, Azam Khan

  2. Dynamic Networks … Time … 2

  3. Year 1 Year 2 3

  4. Egocentric Analysis My co-author’s co-author Me My co-author 4

  5. Egocentric Analysis My co-author’s co-author 2nd level alter Me Ego My co-author 1st level alter 5

  6. Node-link diagrams Animation Matrices [Bach et al., 2013] Projection [Bach et al., 2014] [van den Elzen et al., 2013] 6

  7. Node-link diagrams Animation ❌ Hard to track changes ❌ High cognitive load Matrices [Bach et al., 2013] Projection ❌ Not suitable for sparse dynamic networks ❌ Missing topological information [Bach et al., 2014] [van den Elzen et al., 2013] 7

  8. Dynamic ego-network analysis questions • Joining, leaving, and recurrence of a co-author? • Connectivity to the focal author (ego)? • Splitting and merging of co-author communities (clusters)? • Stability of co-authorships? • … Based on [Lee et al., 2006] and [Ahn et al., 2014] 8

  9. 9

  10. 10

  11. The focal author (ego) 11

  12. A co-author (alter) The focal author (ego) 12

  13. 13

  14. Missing in a year 14

  15. 15

  16. A connection 16

  17. Co-author clusters 17

  18. Co-author clusters 18

  19. 2nd level co-authors 1st level co-authors 19

  20. EgoLines 20

  21. Controlled user study • 18 participants • 13 males and 5 females • 13 analytical tasks (2 categories) • T emporal analysis, topological analysis • 3 techniques • EgoLines (EL), node-links (NL), small multiples (SM) • 1 dataset • IEEE VIS conferences co-authorship networks 21

  22. Small multiples (SM) Node-links (NL) 22

  23. Main take-away 23

  24. Overall > > ~ > Time Temporal analysis > > Accuracy > > Time Topological > ~ Accuracy analysis Finding > > bridges 24

  25. Domain expert evaluation 25

  26. Limitations and future work • More effective overview • Reduce visual clutter • Handle larger ego-networks • Multi-scale aggregation of lines • More experiments • On other datasets and applications • Shed light on facilitating the bridges-finding task 26

  27. Egocentric Analysis of Dynamic Networks with EgoLines Jian Zhao, Michael Glueck, Fanny Chevalier, Yanhong Wu, Azam Khan Contact: jian.zhao@autodesk.com Web: http://jeffjianzhao.github.io/

Recommend


More recommend