Proof about min spanning tree property

Click For Summary
The discussion centers on proving that if an edge e is part of some minimum spanning tree (MST) of a graph G, then it must be the lightest edge in at least one cutset of G. The cut property is highlighted, stating that a lightest edge in a cutset must belong to all MSTs. The original poster considers a proof by contradiction, assuming e is not the lightest edge in any cutset, leading to the conclusion that there exists a heavier edge in every cutset. The conversation suggests exploring specific cutsets or vertex partitions to identify scenarios where e is the lightest edge. The thread concludes with the original poster being banned for multiple accounts, resulting in its closure.
ciphone
Messages
3
Reaction score
0

Homework Statement


(a) If e is part of some MST of G, then it must be a lightest edge in some cutset of G.

Homework Equations


Cut property

The Attempt at a Solution


When the cutset has just one edge then yes it's true obviously. I am think I can do this by contradiction. Assuming e_i is part of some MST of G and e_i is not the lightest edge in any cutset. If this is true then, in every cutset, there exists an edge e_k such that e_k>e_i. I don't know how to go from here

The cut property says For any cut C of the graph, if the weight of an edge in the cut-set of C is strictly smaller than the weights of all other edges of the cut-set of C, then this edge belongs to all MSTs of the graph.

It doesn't say anything about the larger edges in the cut
 
Physics news on Phys.org
It's a bit hard to give help on this without giving the whole thing away. Still, I'll try.

Since the problem says 'for some cutset in G', it seems likely that there are some (possibly many, possibly even most) cutsets containing e in which e is not the lightest edge. So we might like to look for special cutsets (in which e does have the desired property), which is the same as looking for special partitions of the vertices. The other piece of info we have is that e is in a minimal spanning tree (MST), call that T.

So, in the interests of finding a special partition of the vertices, using the info given, can you think of a partition of the vertices that is uniquely defined as a function of only G, T and e, without requiring any additional choices (which would necessarily be arbitrary)?
 
Last edited:
OP has been banned for having multiple accounts, so the thread is now closed.
 

Similar threads

  • · Replies 22 ·
Replies
22
Views
4K
Replies
1
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 13 ·
Replies
13
Views
5K
Replies
3
Views
4K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 13 ·
Replies
13
Views
4K
Replies
2
Views
2K