|Problems? Is your data what you think it is?
Most of what I know about graph theory I have learned here but I don't see why not. On the other hand, what you have just described is an unweighted graph which is stated as a given in my original node. I would be VERY impressed if an algorithm exists that can intuitively know not to follow paths that can lead to dead ends without keeping some sort of a cache of previously followed paths. If they can, then certainly they will be more efficient then what I propose and would love to learn about them.
Cheers - L~R