Scribd
Upload
55 views

Data Structures and Algorithms: A.Baskar

The document discusses data structures and algorithms. It covers tree abstract data types, binary search trees, and AVL trees. AVL trees balance the height between the left and right subtrees, ensuring search, insert, and delete operations take O(log n) time.

Uploaded by

sachnish123
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
55 views

Data Structures and Algorithms: A.Baskar

The document discusses data structures and algorithms. It covers tree abstract data types, binary search trees, and AVL trees. AVL trees balance the height between the left and right subtrees, ensuring search, insert, and delete operations take O(log n) time.

Uploaded by

sachnish123
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 18

Data Structures and Algorithms

A.Baskar
BITS Pilani, K. K. Birla Goa Campus

16 February 2018

A.Baskar (BITS-Pilani) Data Structures and Algorithms 16 February 2018 1 / 10


Recap

Non linear data structures: Trees


Tree ADT
Heap Sort

A.Baskar (BITS-Pilani) Data Structures and Algorithms 16 February 2018 2 / 10


Tree Abstract Data Type

Tree ADT stores elements at nodes


element(v) returns the object stored at the node v O(1)
size(), root(), parent(v)O(1)
children(v) O(cv )
isInternal(v), isExternal(v), IsRoot(v) O(1)
elements(), positions() O(n)
swapElements(v,w), replaceElements(v,e) O(1)

A.Baskar (BITS-Pilani) Data Structures and Algorithms 16 February 2018 3 / 10


Outline

Dictionary ADT
Binary Search Tree
AVL Tree

A.Baskar (BITS-Pilani) Data Structures and Algorithms 16 February 2018 4 / 10


Dictionary Abstract Data Type

Store items (k,e)


Search(k), insert (k,e), remove(k)
size(), isEmpty()

A.Baskar (BITS-Pilani) Data Structures and Algorithms 16 February 2018 5 / 10


Dictionary Abstract Data Type

Store items (k,e)


Search(k), insert (k,e), remove(k)
size(), isEmpty()
Ordered Dictionary ADT: maximum(), minimum(), successor(k),
predecessor(k)

A.Baskar (BITS-Pilani) Data Structures and Algorithms 16 February 2018 5 / 10


Implementations

Logfile, Direct Address table, Hash table


Either they don’t support min, max, succ, pred operations
Or they are not efficient

A.Baskar (BITS-Pilani) Data Structures and Algorithms 16 February 2018 6 / 10


Implementations

Logfile, Direct Address table, Hash table


Either they don’t support min, max, succ, pred operations
Or they are not efficient
Solution is to store data in a non-linear data structures
Store such a way that it can support all these operations

A.Baskar (BITS-Pilani) Data Structures and Algorithms 16 February 2018 6 / 10


Binary Search Tree

Store items in the internal nodes of binary tree


No items in the external nodes

A.Baskar (BITS-Pilani) Data Structures and Algorithms 16 February 2018 7 / 10


Binary Search Tree

Store items in the internal nodes of binary tree


No items in the external nodes
Key of a node is always less than every keys of the nodes in the
left subtree
Key of a node is always greater than every keys of the nodes in
the right subtree

A.Baskar (BITS-Pilani) Data Structures and Algorithms 16 February 2018 7 / 10


Binary Search Tree

Store items in the internal nodes of binary tree


No items in the external nodes
Key of a node is always less than every keys of the nodes in the
left subtree
Key of a node is always greater than every keys of the nodes in
the right subtree
All keys are distinct

A.Baskar (BITS-Pilani) Data Structures and Algorithms 16 February 2018 7 / 10


Performance

search, insert, delete will take O(h) time where h is height of the
tree
But in the worst case h can be O(n)

A.Baskar (BITS-Pilani) Data Structures and Algorithms 16 February 2018 8 / 10


Performance

search, insert, delete will take O(h) time where h is height of the
tree
But in the worst case h can be O(n)
The is because the tree might be unbalanced
We have seen one solution: AVL trees

A.Baskar (BITS-Pilani) Data Structures and Algorithms 16 February 2018 8 / 10


AVL Trees

Balance the height between left subtree and right subtree


Height of the tree is O(log n)

A.Baskar (BITS-Pilani) Data Structures and Algorithms 16 February 2018 9 / 10


AVL Trees

Balance the height between left subtree and right subtree


Height of the tree is O(log n)
All the operations can be done in O(log n)

A.Baskar (BITS-Pilani) Data Structures and Algorithms 16 February 2018 9 / 10


Height Balance Property

For every internal node of a binary search tree, the heights of


the children of v can differ by at most 1
It is a property which characterizes the structure of BST

A.Baskar (BITS-Pilani) Data Structures and Algorithms 16 February 2018 10 / 10


Height Balance Property

For every internal node of a binary search tree, the heights of


the children of v can differ by at most 1
It is a property which characterizes the structure of BST
Any BST which satisfies height balance property is an AVL Tree
(Adelson-Velskii and Landis)

A.Baskar (BITS-Pilani) Data Structures and Algorithms 16 February 2018 10 / 10


Height Balance Property

For every internal node of a binary search tree, the heights of


the children of v can differ by at most 1
It is a property which characterizes the structure of BST
Any BST which satisfies height balance property is an AVL Tree
(Adelson-Velskii and Landis)
Subtree of an AVL tree is an AVL tree
Height of AVL tree stroing is n keys is O(log n)

A.Baskar (BITS-Pilani) Data Structures and Algorithms 16 February 2018 10 / 10

You might also like