Second Largest element in n-ary tree

Given an N-ary tree, find and return the node with second largest value in the given tree. Return NULL if no node with required value is present.

For example, in the given tree

Second largest node is 20.

Recommended: Please try your approach on {IDE} first, before moving on to the solution.

A simple solution is to traverse the array twice. In the first traversal find the maximum value node. In the second traversal find the greatest element node less than the element obtained in first traversal. The time complexity of this solution is O(n).

An Efficient Solution can be to find the second largest element in a single traversal.
Below is the complete algorithm for doing this:

1) Initialize two nodes first and second to NULL as,
   first = second = NULL
2) Start traversing the tree,
   a) If the current node data say root->key is greater
      than first->key then update first and second as,
      second = first
      first = root
   b) If the current node data is in between first and 
      second, then update second to store the value
      of current node as
        second = root
3) Return the node stored in second.

// CPP program to find second largest node 
// in an n-ary tree. 
#include <bits/stdc++.h> 
using namespace std; 
  
// Structure of a node of an n-ary tree 
struct Node { 
    int key; 
    vector<Node*> child; 
}; 
  
// Utility function to create a new tree node 
Node* newNode(int key) 
{ 
    Node* temp = new Node; 
    temp->key = key; 
    return temp; 
} 
  
void secondLargestUtil(Node* root, Node** first, 
                       Node** second) 
{ 
    if (root == NULL) 
        return; 
  
    // If first is NULL, make root equal to first 
    if (!(*first)) 
        *first = root; 
  
    // if root is greater than first then second 
    // will become first and update first equal 
    // to root 
    else if (root->key > (*first)->key) { 
        *second = *first; 
        *first = root; 
    } 
  
    // If root is less than first but greater than 
    // second 
    else if (!(*second) || root->key > (*second)->key) 
        *second = root; 
  
    // number of children of root 
    int numChildren = root->child.size(); 
  
    // recursively calling for every child 
    for (int i = 0; i < numChildren; i++) 
        secondLargestUtil(root->child[i], first, second); 
} 
  
Node* secondLargest(Node* root) 
{ 
    // second will store the second highest value 
    Node* second = NULL; 
  
    // first will store the largest value in the tree 
    Node* first = NULL; 
  
    // calling the helper function 
    secondLargestUtil(root, &first, &second); 
  
    if (second == NULL) 
        return NULL; 
  
    // This handles the case when every element in tree are same. 
    if (first->key == second->key) 
        return NULL; 
  
    // return the second largest element 
    return second; 
} 
  
// Driver program 
int main() 
{ 
    /*   Let us create below tree 
   *             5 
   *         /   |  \ 
   *         1   2   3 
   *        /   / \   \ 
   *       15  4   5   6 
   */
  
    Node* root = newNode(5); 
    (root->child).push_back(newNode(1)); 
    (root->child).push_back(newNode(2)); 
    (root->child).push_back(newNode(3)); 
    (root->child[0]->child).push_back(newNode(15)); 
    (root->child[1]->child).push_back(newNode(4)); 
    (root->child[1]->child).push_back(newNode(5)); 
    (root->child[2]->child).push_back(newNode(6)); 
  
    if (secondLargest(root) != NULL) 
        cout << "Second largest element is : " << secondLargest(root)->key << endl; 
    else
        cout << "Second largest element not found\n"; 
  
    return 0; 
} 

Output:

Second largest element is : 6

This article is contributed by Chhavi. If you like GeeksforGeeks and would like to contribute, you can also write an article using contribute.geeksforgeeks.org or mail your article to contribute@geeksforgeeks.org. See your article appearing on the GeeksforGeeks main page and help other Geeks.

Please write comments if you find anything incorrect, or you want to share more information about the topic discussed above.

My Personal Notes

Recommended: Please try your approach on {IDE} first, before moving on to the solution.

Recommended Posts: