Searching in a Graph

Representations of Graphs

•Adjacency Matrices

•Adjacency Lists
 Adjacency Matrices
Graphs G = (V, E) can be represented by adjacency matrices
G[v1..v|V |, v1..v|V |], where the rows and columns are indexed by the nodes,
and the entries G[vi, vj] represent the edges. In the case of unlabeled
graphs, the entries are just boolean values.

A B C D
A 0 1 1 1

            
B
C
1
1
0
0
0
0
1
1
D 1 1 1 0
 Adjacency Matrices
In case of labeled graphs, the labels themselves may be introduced into
the entries.

A B C D
A 10 4 1

 
 

B     15
 

                  C      9
D        

Adjacency matrices require O(|V |2) space, and so they are space-efficient only when
they are dense (that is, when the graphs have many edges). Time-wise, the adjacency
matrices allow easy addition and deletion of edges.
 Adjacency Lists
A representation of the graph consisting of a list of nodes, with each
node containing a list of its neighboring nodes.

                     
                                 

This representation takes O(|V | + |E|) space.

 Graph Traversal

•Depth-First Traversal

•Breadth-First Traversal
 Depth-First Traversal

• algorithm  dft(x)     
visit(x)     
FOR each y such that (x,y) is an edge DO       
IF <y was not visited yet >
THEN          dft(y) 
 Depth-First Traversal

• A recursive algorithm implicitly recording a

“backtracking” path from the root to the
node currently under consideration
Depth-First Traversal
Depth-First Traversal
Depth-First Traversal
Depth-First Traversal
 Depth-First Traversal
•Depth first search is another way of traversing graphs,
which is closely related to preorder traversal of a tree.

•the Breath-first search tree is typically "short and

bushy", the DFS tree is typically "long and stringy".
 Breadth-First Traversal

Visit the nodes at level i before the nodes of level i+1.

                                       
 Breadth-First Traversal
visit(start node) 
queue <- start node 
WHILE queue is not empty DO    
x <- queue    
FOR each y such that (x,y) is an edge 
                  and y has not been visited yet DO 
    
visit(y)      
queue <- y    
END 
END 
Breadth-First Traversal
Breadth-First Traversal
Breadth-First Traversal
 Breadth-First Traversal
• Each vertex is clearly
• marked at most once,
• added to the list at most once (since that happens only when it's
marked), and
• removed from the list at most once.
Since the time to process a vertex is proportional to the length of its
adjacency list, the total time for the whole algorithm is O(m).

• A tree T constructed by the algorithm is called a breadth first search

tree.

• The traversal goes a level at a time, left to right within a level (where a
level is defined simply in terms of distance from the root of the tree).
 Breadth-First Traversal
• Every edge of G can be classified into one of three groups.
• Some edges are in T themselves.
• Some connect two vertices at the same level of T.
• The remaining ones connect two vertices on two adjacent levels. It
is not possible for an edge to skip a level.
• Breadth-first search tree really is a shortest path tree starting from
its root.
 Relation between BFS and DFS
bfs(G) dfs(G)
{ {
list L = empty tree list L = empty
T = empty tree T = empty
choose a starting vertex x choose a starting vertex x
search(x) search(x)
while(L nonempty) while(L nonempty)
remove edge (v,w) from start of L remove edge (v,w) from end of L
if w not yet visited if w not yet visited
{ {
add (v,w) to T add (v,w) to T
search(w) search(w)
} }
} }

search(vertex v) {
visit(v);
for each edge (v,w)
add edge (v,w) to end of L
}
 Relation between BFS and DFS

Both of these search algorithms now keep a list of edges to

explore; the only difference between the two is

while both algorithms adds items to the end of L,

•BFS removes them from the beginning, which results in

maintaining the list as a queue

•DFS removes them from the end, maintaining the list as a

stack.
 BFS and DFS in directed graphs
The same search routines work essentially unmodified for
directed graphs.
The only difference is that when exploring a vertex v, we
only want to look at edges (v,w) going out of v; we ignore
the other edges coming into v.
For BFS in directed graphs each edge of the graph either
•connects two vertices at the same level
•goes down exactly one level
•goes up any number of levels.
For DFS, each edge either connects an ancestor to
•a descendant
•a descendant to an ancestor
•one node to a node in a previously visited subtree.
Searching
Searching
Searching
Searching
Searching
Searching
Searching
Searching
Searching
End
End