-
- [x] Channel profile -- heatmap and conversation characteristics
- [ ] Network profile -- presence and message exchange social networks, network analysis, hyperedges, thread-level analysis
- [x] User…
-
KaHyPar produces partitions with empty blocks when epsilon is large.
Local search algorithms like 2FM and k-way FM don't do moves if the source block would become
empty. Therefore I assume the proble…
-
In the first cut analysis, let us bunch the To, CC and BCC together; we can form a hyperedge FROM to To, CC, BCC users. Each email gives one hyperedge. We can find about statistical distribution on th…
-
@aschriner Relates to #451 but is more generic
A generic way of autopositioning nodes could apply to several real usecases, like visualising circuits, bio pathways, etc.
- Simulating edge-edge hyperg…
-
Related to issue #287, but importing is a different task.
-
We can model each discussion thread as one hypergraph. All the email header information can be represented as one hyperedge of a hypergraph. A conscise format for representing a discussion thread is a…
-
Or give me libvorbis.a, libvorbisfile.a and libogg.a for Windows because i cannot compile them with MinGW.
-
I think we should get the node degree and visualize it. This is trivial in neo4j:
http://neo4j.com/docs/stable/rest-api-node-degree.html
I think this is important for rankings (e.g., by looking for a…
-
I would like to request a feature for the merging of edges, where two or more edges meet before arriving at the target node.
I currently implement this by adding an extra node that is invisible (width…
-
Hello,
from the documentation, it isn't clear to me whether I am allowed to float_sort a container of `pair` using a comparator that compares the first member and in case of equality disambiguates by…