Home

Decision One

Matchings

A matching is a bi-partite graph in which each vertex is connected to at most one other vertex.

A maximal matching is a matching in which the number of edges is as large as possible.

A complete matching is a matching when there are n vertices in each group, and n edges in the matching.

Alternating Path Algorithm

This algorithm is used for improving bi-partite matchings.

  1. Start with a reasonable [easy to spot] matching
  2. Choose a vertex that isn't in the matching
  3. Draw in edges from the original graph which aren't included in the matching [for this, use a black line]
  4. From these draw in edges which are in the matching [use a red line]
  5. Continue the alternating process until either:
    1. You reach a previously unconnected vertex, or
    2. You get stuck
  6. Include those new edges in your matching. Check if it is maximal, if not repeat algorithm

i) is called a breakthrough. In this case, you swap all the connections [on that arm of the linkage diagram].

ii) means there is no way of improving the current matching.

It is very important that you LABEL and ANNOTATE your matchings, otherwise you will get NO MARKS:

Remember!

You must label "Using the....from vertex C" and "change the status..." everytime you start a new link!
  1. Show initial matching and label "Initial Matching"
  2. Say "Using the Alternate Path Algorithm, starting from vertex A"
  3. Diagram of paths (include the colour key!)
  4. Must have labels of "breakthrough" / "stuck"
  5. (Breakthrough)Say: "Changing the status of the edges" and re-draw
  6. Draw diagram of improved matching