Chapter IX: Binary Trees

Lesson I: General Trees

A graph which has no cycle is called an acyclic graph. A tree is an acyclic graph or graph
having no cycles.

A tree or general trees is defined as a non-empty finite set of elements called vertices or
nodes having the property that each node can have minimum degree 1 and maximum
degree n. It can be partitioned into n+1 disjoint subsets such that the first subset contains
the root of the tree and remaining n subsets includes the elements of the n subtree.

Directed Trees:
A directed tree is an acyclic directed graph. It has one node with indegree 1, while all other
nodes have indegree 1 as shown in fig:
The node which has outdegree 0 is called an external node or a terminal node or a leaf. The
nodes which have outdegree greater than or equal to one are called internal node.

Ordered Trees:
If in a tree at each level, an ordering is defined, then such a tree is called an ordered tree.

Example: The trees shown in the figures represent the same tree but have different orders.
Properties of Trees:
1. There is only one path between each pair of vertices of a tree.
2. If a graph G there is one and only one path between each pair of vertices G is a tree.
3. A tree T with n vertices has n-1 edges.
4. A graph is a tree if and only if it a minimal connected.

Rooted Trees:
If a directed tree has exactly one node or vertex called root whose incoming degrees is 0
and all other vertices have incoming degree one, then the tree is called rooted tree.

Note: 1. A tree with no nodes is a rooted tree (the empty tree)

2. A single node with no children is a rooted tree.
Path length of a Vertex:
The path length of a vertex in a rooted tree is defined to be the number of edges in the path
from the root to the vertex.

Example: Find the path lengths of the nodes b, f, l, q as shown in fig:

Solution: The path length of node b is one.

The path length of node f is two.
The path length of node l is three
The path length of the node q is four.
Lesson II: Binary Trees:
If the outdegree of every node is less than or equal to 2, in a directed tree than the tree is
called a binary tree. A tree consisting of the nodes (empty tree) is also a binary tree. A
binary tree is shown in fig:

Basic Terminology:
Root: A binary tree has a unique node called the root of the tree.

Left Child: The node to the left of the root is called its left child.

Right Child: The node to the right of the root is called its right child.

Parent: A node having a left child or right child or both are called the parent of the nodes.

Siblings: Two nodes having the same parent are called siblings.

Leaf: A node with no children is called a leaf. The number of leaves in a binary tree can
vary from one (minimum) to half the number of vertices (maximum) in a tree.

Descendant: A node is called descendant of another node if it is the child of the node or
child of some other descendant of that node. All the nodes in the tree are descendants of
the root.

Left Subtree: The subtree whose root is the left child of some node is called the left
subtree of that node.

Example: For the tree as shown in fig:

o Which node is the root?

o Which nodes are leaves?
o Name the parent node of each node
Solution: (i) The node A is the root node.
(ii) The nodes G, H, I, L, M, N, O are leaves.
(iii) Nodes Parent
B, C A
D, E B
F C
G, H D
I, J E
K F
L, M J
N, O K

Right Subtree: The subtree whose root is the right child of some node is called the right
subtree of that node.

Level of a Node: The level of a node is its distance from the root. The level of root is
defined as zero. The level of all other nodes is one more than its parent node. The
maximum number of nodes at any level N is 2N.

Depth or Height of a tree: The depth or height of a tree is defined as the maximum
number of nodes in a branch of a tree. This is more than the maximum level of the tree,
i.e., the depth of root is one. The maximum number of nodes in a binary tree of depth d is
2d-1, where d ≥1.

External Nodes: The nodes which have no children are called external nodes or terminal
nodes.
Internal Nodes: The nodes which have one or more than one children are called internal
nodes or non-terminal nodes.

Binary Expression Trees:

An algebraic expression can be conveniently expressed by its expression tree. An expression
having binary operators can be decomposed into
<left operand or expression> (operator) <right operand or expression>

Depending upon precedence of evaluation.

The expression tree is a binary tree whose root contains the operator and whose left
subtree contains the left expression, and right subtree contains the right expression.

Example: Construct the binary expression tree for the expression (a+b)*(d/c)

Solution: The binary expression tree for the expression (a+b)*(d/c) is shown in fig:

Complete Binary Tree: Complete binary tree is a binary tree if it is all levels, except
possibly the last, have the maximum number of possible nodes as for left as possible. The
depth of the complete binary tree having n nodes is log 2 n+1.

Example: The tree shown in fig is a complete binary tree.

Full Binary Tree: Full binary tree is a binary tree in which all the leaves are on the same
level and every non-leaf node has two children.
Differentiate between General Tree and Binary Tree

General Tree Binary Tree

1. There is no such tree having zero 1. There may be an empty binary tree.
nodes or an empty general tree.

2. If some node has a child, then 2. If some node has a child, then it is
there is no such distinction. distinguished as a left child or a right
child.

3. The trees shown in fig are the 3. The trees shown in fig are distinct,
same, when we consider them as when we consider them as binary trees,
general trees. because in (4) is the right child of 2 while
in (ii) 4 is a left child of 2.
Lesson III: Traversing Binary Trees
Traversing means to visit all the nodes of the tree. There are three standard methods to
traverse the binary trees. These are as follows:

1. Preorder Traversal
2. Postorder Traversal
3. Inorder Traversal

1. Preorder Traversal: The preorder traversal of a binary tree is a recursive process. The
preorder traversal of a tree is

o Visit the root of the tree.

o Traverse the left subtree in preorder.
o Traverse the right subtree in preorder.

2. Postorder Traversal: The postorder traversal of a binary tree is a recursive process.

The postorder traversal of a tree is

o Traverse the left subtree in postorder.

o Traverse the right subtree in postorder.
o Visit the root of the tree.

3. Inorder Traversal: The inorder traversal of a binary tree is a recursive process. The
inorder traversal of a tree is

o Traverse in inorder the left subtree.

o Visit the root of the tree.
o Traverse in inorder the right subtree.

Example: Determine the preorder, postorder and inorder traversal of the binary tree as
shown in fig:
Solution: The preorder, postorder and inorder traversal of the tree is as follows:

Preorder 1 2 3 4 5 6 7 8 9 10 11

Postorder 3 5 4 2 7 10 9 11 8 6 1

Inorder 3 2 5 4 1 7 6 9 10 8 11

Algorithms:
(a)Algorithm to draw a Unique Binary Tree when Inorder and Preorder Traversal
of the tree is Given:

1. We know that the root of the binary tree is the first node in its preorder. Draw the
root of the tree.
2. To find the left child of the root node, first, use the inorder traversal to find the
nodes in the left subtree of the binary tree. (All the nodes that are left to the root
node in the inorder traversal are the nodes of the left subtree). After that, the left
child of the root is obtained by selecting the first node in the preorder traversal of
the left subtree. Draw the left child.
3. In the same way, use the inorder traversal to find the nodes in the right subtree of
the binary tree. Then the right child is obtained by selecting the first node in the
preorder traversal of the right subtree. Draw the right child.
 4. Repeat the steps 2 and 3 with each new node until every node is not visited in the
preorder. Finally, we obtain a unique tree.

Example: Draw the unique binary tree when the inorder and preorder traversal is given as
follows:

Inorder B A D C F E J H K G I

Preorder A B C D E F G H J K I

Solution: We know that the root of the binary tree is the first node in preorder traversal.
Now, check A, in the inorder traversal, all the nodes that are of left A, are nodes of left
subtree and all the nodes that are right of A, are nodes of right subtree. Read the next node
in the preorder and check its position against the root node, if its left of the root node, then
draw it as a left child, otherwise draw it a right child. Repeat the above process for each
new node until all the nodes of the preorder traversal are read and finally we obtain the
binary tree as shown in fig:

(b) Algorithm to draw a Unique Binary Tree when Inorder and Postorder Traversal
of the tree is Given:

1. We know that the root of the binary tree is the last node in its postorder. Draw the
root of the tree.
 2. To find the right child of the root node, first, use the inorder traversal to find the
nodes in the right subtree of the binary tree. (All the nodes that are right to the root
node in the inorder traversal are the nodes of the right subtree). After that, the right
child of the root is obtained by selecting the last node in the postorder traversal of
the right subtree. Draw the right child.
3. In the same way, use the inorder traversal to find the nodes in the left subtree of the
binary tree. Then the left child is obtained by selecting the last node in the postorder
traversal of the left subtree. Draw the left child.
4. Repeat the steps 2 and 3 with each new node until every node is not visited in the
postorder. After visiting the last node, we obtain a unique tree.

Example: Draw the unique binary tree for the given Inorder and Postorder traversal is
given as follows:

Inorder 4 6 10 12 8 2 1 5 7 11 13 9 3

Postorder 12 10 8 6 4 2 13 11 9 7 5 3 1

Solution: We know that the root of the binary tree is the last node in the postorder
traversal. Hence, one in the root node.

Now, check the inorder traversal, we know that root is at the center, hence all the nodes
that are left to the root node in inorder traversal are the nodes of left subtree and, all that
are right to the root node are the nodes of the right subtree.

Now, visit the next node from back in postorder traversal and check its position in
inorder traversal, if it is on the left of root then draw it as left child and if it is on the right,
then draw it as the right child.

Repeat the above process for each new node, and we obtain the binary tree as shown in fig:
(c)Algorithm to convert General Tree into the binary tree

1. Starting from the root node, the root of the tree is also the root of the binary tree.
2. The first child C1(from left) of the root node in the tree is the left child C1 of the root
node in the binary tree, and the sibling of the C 1 is the right child of C1 and so on.
3. Repeat the step2 for each new node.

Example: Convert the following tree as shown in fig into a binary tree.
Solution: The root of the tree is the root of the binary tree. Hence A is the root of the
binary tree. Now B becomes the left child of A in a binary tree, C becomes the right child of
B, D becomes right child of C, and E becomes the right child of D in the binary tree and
similarly applying the algorithm we obtain the binary tree as shown in fig:
Lesson IV: Binary Search Trees
Binary search trees have the property that the node to the left contains a smaller value than
the node pointing to it and the node to the right contains a larger value than the node
pointing to it.

It is not necessary that a node in a 'Binary Search Tree' point to the nodes whose value
immediately precede and follow it.

Example: The tree shown in fig is a binary search tree.

Inserting into a Binary Search Tree: Consider a binary tree T. Suppose we have given
an ITEM of information to insert in T. The ITEM is inserted as a leaf in the tree. The
following steps explain a procedure to insert an ITEM in the binary search tree T.

1. Compare the ITEM with the root node.

2. If ITEM>ROOT NODE, proceed to the right child, and it becomes a root node for the
right subtree.
3. If ITEM<ROOT NODE, proceed to the left child.
4. Repeat the above steps until we meet a node which has no left and right subtree.
5. Now if the ITEM is greater than the node, then the ITEM is inserted as the right child,
and if the ITEM is less than the node, then the ITEM is inserted as the left child.

Example: Show the binary search tree after inserting 3, 1,4,6,9,2,5,7 into an initially
empty binary search tree.

Solution: The insertion of the above nodes in the empty binary search tree is shown in fig:
Deletion in a Binary Search Tree: Consider a binary tree T. Suppose we want to delete a
given ITEM from binary search tree. To delete an ITEM from a binary search tree we have
three cases, depending upon the number of children of the deleted node.

1. Deleted Node has no children: Deleting a node which has no children is very
simple, as replace the node with null.
2. Deleted Node has Only one child: Replace the value of a deleted node with the
only child.
3. Deletion node has only two children: In this case, replace the deleted node with
the node that is closest in the value to the deleted node. To find the nearest value,
we move once to the left and then to the right as far as possible. This node is called
the immediate predecessor. Now replace the value of the deleted node with the
immediate predecessor and then delete the replaced node by using case1 or case2.

Example: Show that the binary tree shown in fig (viii) after deleting the root node.

Solution: To delete the root node, first replace the root node with the closest elements of
the root. For this, first, move one step left and then to the right as far as possible to the
node.Then delete the replaced node. The tree after deletion shown in fig:
Lesson V: Spanning Tree
A subgraph T of a connected graph G is called spanning tree of G if T is a tree and T include
all vertices of G.

Minimum Spanning Tree:

Suppose G is a connected weight graph i.e., each edge of G is assigned a non-negative
number called the weight of edge, then any spanning tree T of G is assigned a total weight
obtained by adding the weight of the edge in T.

A minimum spanning tree of G is a tree whose total weight is as small as possible.

Kruskal's Algorithm to find a minimum spanning tree: This algorithm finds the
minimum spanning tree T of the given connected weighted graph G.

1. Input the given connected weighted graph G with n vertices whose minimum
spanning tree T, we want to find.
2. Order all the edges of the graph G according to increasing weights.
3. Initialize T with all vertices but do include an edge.
4. Add each of the graphs G in T which does not form a cycle until n-1 edges are added.

Example1: Determine the minimum spanning tree of the weighted graph shown in fig:
Solution: Using kruskal's algorithm arrange all the edges of the weighted graph in
increasing order and initialize spanning tree T with all the six vertices of G. Now start adding
the edges of G in T which do not form a cycle and having minimum weights until five edges
are not added as there are six vertices.

Edges Weights Added or Not

(B, E) 2 Added
(C, D) 3 Added
(A, D) 4 Added
(C, F) 4 Added
(B, C) 5 Added
(E, F) 5 Not added
(A, B) 6 Not added
(D, E) 6 Not added
(A, F) 7 Not added

Step1:

Step2:

Step3:
Step4:

Step5:

Step6: Edge (A, B), (D, E) and (E, F) are discarded because they will form the cycle in a
graph.

So, the minimum spanning tree form in step 5 is output, and the total cost is 18.
Example2: Find all the spanning tree of graph G and find which is the minimal spanning
tree of G shown in fig:

Solution: There are total three spanning trees of the graph G which are shown in fig:
To find the minimum spanning tree, use the KRUSKAL'S ALGORITHM. The minimal spanning
tree is shown in fig:

Edges Weights Added or Not

(E, F) 1 Added
(A, B) 2 Added
(C, D) 2 Added
(B, C) 3 Added
(D, E) 3 Added
(B, D) 6 Not Added

The first one is the minimum spanning having the minimum weight = 11.