ManyNets: Visualize Many Networks Simultaneously
Project Description
ManyNets is a network visualization tool with tabular interface designed to visualize up to several thousand network overviews at once. This allows networks to be compared, and large networks to be explored using a divide-and-conquer approach. For example, comparing different social networks can provide insights into the underlying causes for their differences. Or an individual social network can also be subdivided into temporal slices, which can then be examined to locate temporal patterns or regions and periods change. Networks can also be subdivided and compared based on motifs (small patterns of connectivity), clusters, or network-specific attributes.
Features
A collection of networks is presented in a table, where each row represents a single network. Columns represent statistics, such as link count, degree distribution, or clustering coefficients. Details are available on-demand, either in the form of larger views or as SocialAction views of specific networks.
 Selecting columns |  Sorting and ranking with a custom query |  Details of a distribution. The green portion corresponds to selected rows. |
The use of a table allows easy comparisons between rows (networks) and columns (their statistics). Computationally expensive statistics (e.g.: network diameter) are only added if the user explicitly requests them. Users can also add custom columns by entering expressions in Python. A similar interface can be used to specify expressions for filters, selections, and sorting column definitions.
Column summaries, placed on top of the column headers, provide abstracts of the contents of their columns. The summaries also support direct user interaction, and reflect application state by highlighting values that correspond to currently-selected rows. The histograms used in these summaries provide quick assessment of the distributions of values within a set of networks.
Video Demonstrations
Name |
Description |
Format |
|
| Short demo | Short demonstration of ManyNets (under 6 minutes) | mp4 | |