Presentation is loading. Please wait.

Presentation is loading. Please wait.

Watson Innovations Cognitive Visualization Lab Cody Dunne ibm.biz/cogvislab August 10, 2015 Graph Summit 2015 Readability metric feedback.

Similar presentations


Presentation on theme: "Watson Innovations Cognitive Visualization Lab Cody Dunne ibm.biz/cogvislab August 10, 2015 Graph Summit 2015 Readability metric feedback."— Presentation transcript:

1 Watson Innovations Cognitive Visualization Lab Cody Dunne cdunne@us.ibm.com ibm.biz/cogvislab August 10, 2015 Graph Summit 2015 Readability metric feedback for aiding node-link visualization designers

2 Cody Dunne, PhD – Cognitive Visualization RSM Web: ibm.biz/codydunneEmail: cdunne@us.ibm.com epidemiology/dynamic networks layout readabilityexploration provenancenetwork type overviewsgroup/set visualization aggregation techniquesliterature explorationnews term occurrence computer network traffic

3 Watson Graph Readability The team Daniel Weidele Research Intern CVL Steve Ross STSM CVL Mauro Martino Manager CVL Ben Shneiderman Professor UMD

4 Why Visualization? Anscombe’s quartet – Table IIIIIIIV xyxyxyxy 10.008.0410.009.1410.007.468.006.58 8.006.958.008.148.006.778.005.76 13.007.5813.008.7413.0012.748.007.71 9.008.819.008.779.007.118.008.84 11.008.3311.009.2611.007.818.008.47 14.009.9614.008.1014.008.848.007.04 6.007.246.006.136.006.088.005.25 4.004.264.003.104.005.3919.0012.50 12.0010.8412.009.1312.008.158.005.56 7.004.827.007.267.006.428.007.91 5.005.685.004.745.005.738.006.89

5 Why Visualization? Anscombe’s quartet – Statistics & Visualization Property in Each Case ValueEquality Mean of x9Exact Variance of x11Exact Mean of y7.50 2 decimal places Variance of y 4.122 or 4.127 3 decimal places Correlation between x & y 0.816 3 decimal places Linear regression line y = 3.00 + 0.500x 2 & 3 decimal places

6 No catalogue of techniques can convey a willingness to look for what can be seen, whether or not anticipated. Yet this is at the heart of exploratory data analysis.... the picture-examining eye is the best finder we have of the wholly unanticipated. – Tukey, 1980 Why Visualization? Tukey

7 Node 1Node 2 AliceBob AliceCathy Alice Node-Link Graph Visualization General Graph ≈ Network Node ≈ Vertex ≈ Entity Edge ≈ Link ≈ Relationship ≈ Tie

8 Watson Graph Readability Comparing two popular layout algorithms D3.js Force LayoutGraphViz SFDP

9 Hachul & Jünger, 2006 Watson Graph Readability Immense variation in layout readability and speed

10 How much of the underlying network structure can you understand from a given layout? Watson Graph Readability Evaluate, compare, and improve layouts Edge CrossingsEdge Crossing Angle Node Overlap

11 Watson Graph Readability Measuring Readability Source: Sugiyama, 2002, p. 14 Simple rules or heuristics Davidson & Harel, 1996 Global readability metrics Purchase, 2002 User performance Huang et al., 2007, etc.

12 Existing metrics New Our metrics Local Node overlap Edge tunnel Drawing space used Group overlap Distance Coherence Edge crossing Angular resolution Edge crossing angle Stress

13

14 Global readability metric [0,1] where: 0 = Complete overlap 1 = No overlap Node readability metric Ratio of node area that overlaps other nodes Watson Graph Readability Node Overlap RM

15 Global readability metric [0,1] where: 0 = All possible crossings 1 = No crossings Edge readability metric Just like gobal RM Watson Graph Readability Edge Crossing RM

16 Node readability metric [0,1] where: 0 = All possible crossings 1 = No crossings Watson Graph Readability Edge Crossing RM (continued)

17 Evaluate, compare, and improve layouts Layout algorithm heuristics and parameters User-generated or user-modified layouts Manual layout suggestions a la snap-to-grid Fully automatic layouts Recommend layouts and parameterizations Watson Graph Readability Goal

18 Watson Graph Readability Layout algorithm & design comparison interface

19 Train a model M(G,S(G),L,P(L))->(RM,UO) Graph G with statistics S(G), layout algorithm L and parameters P(L) Readability metrics for L on G with P(L) argmax_{L,P(L)|(G?),S(G),RM'} M, with RM' ⊆ RM as the optimal layout Watson Graph Readability Machine learning to identify best layout

20 Need interface on top of your graph store – Data cleaning, process sanity check – Exploration Must be able to evaluate effectiveness Works with aggregate views Watson Graph Readability Use in practice

21 Raise awareness of readability issues Localized identification of where improvement is needed Optimization recommendations for tasks Interactive optimization Future optimization plans Dunne C, Ross SI, Shneiderman B, and Martino M (2015), “Readability metric feedback for aiding node-link visualization designers”. IBM Journal of Research and Development. Dunne C and Shneiderman B (2009), "Improving graph drawing readability by incorporating readability metrics: A software tool for network analysts". University of Maryland. Human-Computer Interaction Lab Tech Report No. (HCIL-2009-13). Watson Graph Readability Discussion

22 Project Information OrganizationWatson Team Size4 Path to MarketResearch Asset IBM S/WSoftLayer, Bluemix Open SourceJava Topology Suite and/or JavaScript Topology Suite, Java EE, MySQL, EmpireDB, JUNG, JQuery

23

24

25

26

27

28

29

30


Download ppt "Watson Innovations Cognitive Visualization Lab Cody Dunne ibm.biz/cogvislab August 10, 2015 Graph Summit 2015 Readability metric feedback."

Similar presentations


Ads by Google