Scribd
Upload
30 views2 pages

DFS and BFS - Non-Recursive Versions

einstein-newtonian method for simulation algorithm to get a nobel prize.The laser characteristics in algo strusture has ushered a new era in neural network modeling.

Uploaded by

Tamim Hossain
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
30 views2 pages

DFS and BFS - Non-Recursive Versions

einstein-newtonian method for simulation algorithm to get a nobel prize.The laser characteristics in algo strusture has ushered a new era in neural network modeling.

Uploaded by

Tamim Hossain
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

DFS and BFS – non-recursive versions

DFS(G, v0 )
Input: graph G and starting vertex v0
Output: labels L(v) for each vertex v that give the order in which the search visits the
vertices. (Each vertex label is initially zero, with a zero label designating an unvisited
vertex.)

• T ←− {v0 } (T is maintained as a stack)

• For all vertices v, L(v) ←− 0.

• Count ←− 1.

• L(v0 ) ←− Count

• While T 6= ∅ do

– v ←− top of stack T
– If there is a vertex w adjacent to v with L(w) = 0 then do
∗ Count ←− Count + 1
∗ L(w) ←− Count
∗ Push w onto stack T
– Else pop v from stack T

BFS(G, v0 )
Input: graph G and starting vertex v0
Output: labels L(v) for each vertex v that give the length of the shortest path from that
vertex to v0 .

• T ←− {v0 } (T is maintained as a queue)

• For all vertices v, L(v) ←− 0.

• L(v0 ) ←− 0

• While T 6= ∅ do

– v ←− first item on queue T


– Remove v from T
– For all w adjacent to v with L(w) = 0 and w 6= v0 do
∗ add w to queue T
∗ L(w) ←− L(v) + 1
Efficiency analysis of DFS:

• We’ll measure input size by the number n of vertices in G.

• The basic operation will be the label check implicit in the line “If there is a vertex w
adjacent to v with L(w) = 0 then . . . ”

• There are at most n − 1 of these label checks each time through the “While . . . ” loop.

• But the “While . . . ” loop is executed at most 2n times: each time either adds a vertex
to the stack or takes on off, and each vertex is added once and taken off once.

• This makes the total number of basic steps no more than n · 2n, so the algorithm has
efficiency in O(n2 ).

Efficiency analysis of BFS:

• Again, we’ll measure input size by the number n of vertices in G.

• The basic operation is again a label check, this time implicit in the line “For all w
adjacent to v with L(w) = 0 do . . . ”

• There are at most n − 1 label checks each time through the “While . . . ” loop.

• But the “While . . . ” loop is executed at most n times, since each time one vertex is
removed from the queue.

• This makes the total number of basic steps no more than n2 , so the algorithm has
efficiency in O(n2 ).

Note: Can you see how the label checks would be handled differently depending on if G is
input as an adjacency matrix or as adjacency lists? Is one better than the other?

You might also like