Gradoop Integration for KNIME Workbench version 1.0.0.v201911212144 by Steffen Dienst
Computes a hierarchical clustering of a logical graph containing 2 levels. The hierarchy levels are written as separate logical graphs to the given path, each to a folder starting with 'level_', e.g. 'level_0' for the lowest level. The top level should have significantly fewer elements than the original graph, which helps to visually explore very big graphs with the Big Graph Viewer node.
The superordinate hierarchy level consists of super vertices, where each is a parent for a number of vertices from the next subordinate hierarchy level and holds references to its children. Vertices in a subordinate level are first assigned to a cluster. Vertices in a cluster are than grouped by label to preserve both, the structure and the semantics of the graph. Each group then forms a super vertex. All edges between super vertices are combined for each direction.
How well the top level elements are reduced depends on the original graph structure and the type of cluster assignment algorithm.
How the vertices are assigned to a cluster / community:
a) Label Propagation: the Flink Gelly Label Propagation algorithm
b) Structure Based Propagation: a custom Label Propagation algorithm, which uses structural properties. Computational more demanding
c) Group By Label: groups all vertices by label for the whole graph. Builds small top level but destroys local graph structures like communities.
To use this node in KNIME, install Gradoop-Integration from the following update site:
You don't know what to do with this link? Read our NodePit Product and Node Installation Guide that explains you in detail how to install nodes to your KNIME Analytics Platform.
You want to see the source code for this node? Click the following button and we’ll use our super-powers to find it for you.
Do you have feedback, questions, comments about NodePit, want to support this platform, or want your own nodes or workflows listed here as well? Do you think, the search results could be improved or something is missing? Then please get in touch! Alternatively, you can send us an email to firstname.lastname@example.org, follow @NodePit on Twitter, or chat on Gitter!
Please note that this is only about NodePit. We do not provide general support for KNIME — please use the KNIME forums instead.