Re: shortest path in bipartite directed graph
- From: Proginoskes <CCHeckman@xxxxxxxxx>
- Date: Fri, 02 Nov 2007 09:41:39 -0000
On Nov 1, 10:09 pm, klochner <kloch...@xxxxxxxxx> wrote:
Are there any specialized algorithms for shortest paths in a
bipartite graph?
Specifically, I have a bipartite graph with two directed
edges between each pair of nodes, with d(x,y) = -d(y,x),
and I want to compute all-pairs shortest paths. I was
going to use floyd-warshall but thought there may be a
better algorithm out there.
I'm not so sure you can use F-W if you have negative distances. (Or
Dijkstra's algorithm, either.) Directed cycles with a total distance
of 0 would at least slow these algorithms down.
You have to be careful about negative distances, because a directed
cycles with a negative total sum. If you have one of those, then there
is no algorithm, because the minimum distance is negative infinity,
and you need to trace an infinite number of edges to find this out.
I would then consider the properties of your distance function (on the
edges), and try to make an algorithm based on those.
--- Christopher Heckman
.
- Follow-Ups:
- Re: shortest path in bipartite directed graph
- From: klochner
- Re: shortest path in bipartite directed graph
- References:
- shortest path in bipartite directed graph
- From: klochner
- shortest path in bipartite directed graph
- Prev by Date: Re: Third dimension...
- Next by Date: Re: Primitive polynomials over GF(2^m)
- Previous by thread: shortest path in bipartite directed graph
- Next by thread: Re: shortest path in bipartite directed graph
- Index(es):
Relevant Pages
|