MultiNet.js A visualization framework for large, dynamic and multi-layered graphs developed by the Chair of Systems Design at ETH Zurich

Features

Navigate through the graph by selecting single or node multiple nodes on the graph or on the data view. 1- and 2-order neighborhoods will highlighted in all layers

Navigation

Filter edges based on their timestamps and replay the evolution of the network by selecting a time window and using the player

Replaying

A perspective and an orthographic camera are available for 3D navigation. 2D navigation is also possible where all layers are rendered side by side.

Camera Options

Each visualization generates a specific url which returns the cached version of the graph which can be copied and distributed.

Sharing

Edit the color of each layer and scale nodes wrt. different graph properties to get different visual results.

Styling

Take a snapshot of the graph at any time preserving the camera perspective and the time filters applied.

Snapshot

Examples


Getting Started

Fill in the form below to receive an active URL.

With your access URL, upload an edgelist with four colums where the header is source;target;layer;timestamp

layer stands for the identifier of the layer which the edge belongs to. The tool supports arbitrary number of layers.

timestamp can be a date string in YYYY-MM-DD format, or just years

Once an edgelist is uploaded, the rendering can be started immediately. But there are more options:

Upload node attributes Uploading node ids used in the edgelist and an arbitrary number of attributes, separated by a ; will allow you to navigate through the nodes from within a formatted table.

Select layout The Fruchterman-Reingold is the default and it is the parallel version of the standard layout algorithm in the igraph package.

Choose directedness

I agree to the terms and conditions


Chair of Systems Design is not liable for any unexpected exposure of the resulting plots. Chair of Systems Design will not store or make use of any of the uploaded data. The user data can be deleted on the user's request, or within 24 hours after the upload