Examples using DyNetVis - Temporal Network Analysis
Some temporal visualizations from different networks, using techniques such as Recurrent Neighbors (RN), Temporal Activity Map (TAM) and Community-based Node Ordering (CNO).
All images were generated in the DyNetVis System.
L. Isella, J. Stehl e, A. Barrat, C. Cattuto, J.-F. Pinton, W. Van den Broeck, What's in a crowd? Analysis of face-to-face behavioral networks. J. Theor. Biol. 271 166-180 (2011).
- Initial temporal visualization (Massive Sequence View - MSV) with Appearance ordering of nodes and default colors in the DyNetVis system:
2. MSV with Appearance ordering and colored nodes and edges:
3. MSV with Recurrent Neighbors ordering and colored nodes and edges:
4. MSV with Community-based node ordering (CNO) and colored nodes and edges, only showing the Intra edges:
5. MSV with Community-based node ordering (CNO) and different color for the nodes and edges, using the Temporal Activity Map (TAM):
6. TAM with black background, highlighting the nodes activity:
7. Coordination between layouts, selecting some nodes from a node-link diagram perspective and the respective nodes and links in the MSV:
L. Isella, M. Romano, A. Barrat, C. Cattuto, V. Colizza, W. Van den Broeck, F. Gesualdo, E. Pandol , L. Rav a, C. Rizzo, A.E. Tozzi, Close encounters in a pediatric ward: measuring face-to-face proximity and mixing patterns with wearable sensors. PLoS ONE 6(2): e17144 (2011).
MSV layout, with CNO (Infomap, RN, RN) in level 1, only showing Intra edges:
- The colors represent the metadata for each profile: Adminstration (blue), Medical (green), Nurse (red), Patient (orange).
MSV layout, with CNO (Louvain, RN, RN) in level 1, only showing Intra edges:
- The colors represent the metadata for each profile: Adminstration (blue), Medical (green), Nurse (red), Patient (orange).
TAM layout, with CNO (Louvain, RN, RN):
- The colors represent the metadata for each profile: Adminstration (blue), Medical (green), Nurse (red), Patient (orange).
- The intensity of the color represent the amount of activity in that timestamp, i.e., higher intensity means more connections of that node in that timestamp.
Pereira, FS, Amo, Sd, Gama, J. Detecting Events in Evolving Social Networks through Node Centrality Analysis. In: Workshop on Largescale Learning from Data Streams in Evolving Environments of ECML PKDD. p. 83–93. (2016).
- MSV layout, from left to right:
- MSV layout using CNO (Louvain, RN, RN) level 1 without the edge filtering
- MSV layout with CNO(Louvain, RN, RN) level 1 showing only intra edges
- MSV layout with CNO(Louvain, RN, RN) level 2 showing only intra edges
- Communities with different patterns
This is a scalability solution to very large networks. See more about CNO.
Museum and Hospital networks can be found at: http://www.sociopatterns.org/