Presentation is loading. Please wait.

Presentation is loading. Please wait.

Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Matej Novotný Comenius University Bratislava, Slovakia Helwig Hauser VRVis Research.

Similar presentations


Presentation on theme: "Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Matej Novotný Comenius University Bratislava, Slovakia Helwig Hauser VRVis Research."— Presentation transcript:

1 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Matej Novotný Comenius University Bratislava, Slovakia Helwig Hauser VRVis Research Center Vienna, Austria

2 Matej Novotný http://www.VRVis.at/ 2 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Our goal A parallel coordinates visualization that: Employs Focus+Context Handles outliers Renders effectively

3 Matej Novotný http://www.VRVis.at/ 3 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Overview Motivation Abstraction, Focus+Context Outliers Workflow Binning Context Benefits Bonus! Results and conclusions

4 Matej Novotný http://www.VRVis.at/ 4 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Parallel Coordinates Insight into multidimensional data Correlations, Groups, Outliers

5 Matej Novotný http://www.VRVis.at/ 5 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Parallel Coordinates Insight into multidimensional data Correlations, Groups, Outliers

6 Matej Novotný http://www.VRVis.at/ 6 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Parallel Coordinates Insight into multidimensional data Correlations, Groups, Outliers

7 Matej Novotný http://www.VRVis.at/ 7 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Large data visualization Large data cause clutter in visualization 16.000 records

8 Matej Novotný http://www.VRVis.at/ 8 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Large data visualization Transparency used to decrease clutter 16.000 records

9 Matej Novotný http://www.VRVis.at/ 9 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Large data visualization Transparency used to decrease clutter ? 32.000 records

10 Matej Novotný http://www.VRVis.at/ 1010 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Large data visualization Transparency used to decrease clutter ?? 64.000 records

11 Matej Novotný http://www.VRVis.at/ 11 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Large data visualization Transparency used to decrease clutter ??? 100.000 records

12 Matej Novotný http://www.VRVis.at/ 1212 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Large data visualization Transparency used to decrease clutter ??? Do these records belong to the main trend?

13 Matej Novotný http://www.VRVis.at/ 1313 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Data abstraction Density-based representation of data Trends are clearly visible 16 bins

14 Matej Novotný http://www.VRVis.at/ 1414 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Related work Hierarchical Parallel Coordinates (Fua et al., 1999) Visual representation of clusters Smooth transparency Cluster centers emphasized

15 Matej Novotný http://www.VRVis.at/ 1515 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Related work Revealing Structure within Clustered Parallel Coordinates Displays (Johansson et al., 2005) Textures, density Transfer functions Clusters Outliers

16 Matej Novotný http://www.VRVis.at/ 1616 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Outliers Different, sparse, rare Why should we care? Investigation (special cases in simulations…) Security (network intrusion, suspicious activity…) Detect errors in data acquisition Outliers can be considered in: Data space Screen space

17 Matej Novotný http://www.VRVis.at/ 1717 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Outliers Outliers are like kids. If you leave them unattended they either get lost or they break stuff.

18 Matej Novotný http://www.VRVis.at/ 1818 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Outliers Avoid losing them in visualization e.g. due to transparency or abstraction Improve data abstraction or F+C e.g. remove outliers from clustering

19 Matej Novotný http://www.VRVis.at/ 1919 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Workflow

20 Matej Novotný http://www.VRVis.at/ 2020 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Workflow

21 Matej Novotný http://www.VRVis.at/ 2121 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Step 1: Binning 2D binning Density-based rep. Screen-oriented Low memory demands compared to nD Every segment separately Result = bin map

22 Matej Novotný http://www.VRVis.at/ 2 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Benefits of binning? Operations no longer depend on the size of the input Information is preserved Variable precision of binning Variable precision of visual output Fine binning does not destroy details Larger bins can be produced from finer bins 128x128 bins

23 Matej Novotný http://www.VRVis.at/ 2323 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Step 2: Outlier detection Various criteria can be employed e.g. isolated bins, median filter … 64x64 bin map32x32 bin map median filter 32x32 bin map isolated bins

24 Matej Novotný http://www.VRVis.at/ 2424 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Step 3: Generating Context Outliers opaque lines Binned trends quads Population mapped to color intensity No blending Low visual complexity Rendering order according to population 8 bins

25 Matej Novotný http://www.VRVis.at/ 2525 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Step 4: Add Focus 8 bins

26 Matej Novotný http://www.VRVis.at/ 2626 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Benefits Operations performed on bin maps Reduced complexity Results coherent with visual output More operations feasible – e.g. Clustering Outliers handled separately Increased information value Clearer context Output-sensitive implementation View divided into layers and segments

27 Matej Novotný http://www.VRVis.at/ 2727 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Results Large data can be rendered and explored 3 millions records, 16 dimensions, 32 bins Binned in 30 sec, rendered instantly (3Ghz,64bit)

28 Matej Novotný http://www.VRVis.at/ 2828 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates BONUS! Clustering

29 Matej Novotný http://www.VRVis.at/ 2929 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Clustering, step 0 Apply Gaussian to smooth out the bin map Segmentation data, Green vs Darkness

30 Matej Novotný http://www.VRVis.at/ 3030 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Clustering, further steps Start with the highest population Decrease the population threshold Old clusters grow New clusters emerge 50%20%10%0%

31 Matej Novotný http://www.VRVis.at/ 3131 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Clustering results R B G D S H

32 Matej Novotný http://www.VRVis.at/ 3232 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Clustering results R B G D S H

33 Matej Novotný http://www.VRVis.at/ 3 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Clustering results R B G D S H

34 Matej Novotný http://www.VRVis.at/ 3434 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Clustering results R B G D S H

35 Matej Novotný http://www.VRVis.at/ 3535 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Conclusions Data abstraction based on density rep. Data operations - outlier detection, clustering Focus+Context Variable context precision Outliers preserved Much clearer view for large data Screen-oriented and output-sensitive Interactive visualization of large data

36 Matej Novotný http://www.VRVis.at/ 3636 Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Acknowledgements K-Plus Vega grant 1/3083/06. AVL List GmbH - data Juergen Platzer Prof. Peter Filzmoser Harald Piringer Michael Wohlfahrt

37 Thank you for your attention!


Download ppt "Outlier-Preserving Focus+Context Visualization in Parallel Coordinates Matej Novotný Comenius University Bratislava, Slovakia Helwig Hauser VRVis Research."

Similar presentations


Ads by Google