Presentation is loading. Please wait.

Presentation is loading. Please wait.

Visualization of Content Information in Networks using GlyphNet

Similar presentations


Presentation on theme: "Visualization of Content Information in Networks using GlyphNet"— Presentation transcript:

1 Visualization of Content Information in Networks using GlyphNet
Anne Denton and Paul Juell Department of Computer Science North Dakota State University Fargo, ND, USA

2 Information Visualization on a Graph
Genomics Protein-protein interactions Biochemical pathways WWW Link structure Scientific publications Citations Social networks Scientific American 05/03

3 Graph/Tree Visualization Tools
Euler 1736 Since then Many layouts Aesthetic rules Navigation Probe capability

4 Navigation / Probe Capability

5 Visual Data Mining as Hypothesis Generation Process
Hypotheses regarding Visualization tools Edge distribution Relationship between node content and edge distribution Relationship between node content of different nodes: Mining relational data ? Graph visualization tool ~~~~ Probe capability GlyphNet

6 Visualization of Node Data??
So far mostly connectivity Exceptions (SGI Filesystem browser) Color Size Glyphs in Clustering P. Eades, Q. Feng, “Multilevel Visualization of Clustered Graphs,” Lecture Notes in Computer Science”, 1190, pp , 1997

7 Glyphs Weather map symbols Chernoff faces
Chernoff, H., “The use of faces to represent points in k-dimensional space graphically,”  Journal of the American Statistical Association, Vol. 68, pp , 1973.

8 Adapting a Star Plot Star plot Our solution
Star with n arms for n attributes Value: distance from center Connect points Our solution Embed in circle Filled pie slices

9 Bioinformatics Example
Motivated by KDD-cup 02 and other bioinformatics problems Graph: Protein-protein interactions in yeast From experiments Undirected graph Categorical and continuous attributes Essential (organism survives gene-deletion experiment Color Red: AHR Green: not AHR Yellow: “control” *AHR: Aryl Hydrocarbon Receptor Signaling Pathway

10 Questions Prediction of red (AHR) What are interesting patterns?
Which attributes in neighbors are relevant? How should we integrate neighbor knowledge? What are interesting patterns? Which properties say more about neighboring nodes than about the node itself? But not:

11 Integration of Results into Other Data Mining Algorithms
Include additional attribute for each object Count-based: Number of neighbors with property “essential” (example: 2) Truth-value-based: Existence of a neighbor with property “essential” (example: true)

12 Summary of GlyphNet Idea: Visualization of node data as glyph
Goal: Identify patterns that involve multiple nodes Next step: Validate pattern numerically Possible use: Include in node-based data mining algorithm


Download ppt "Visualization of Content Information in Networks using GlyphNet"

Similar presentations


Ads by Google