Open In App

Count of subarrays whose maximum element is greater than k

Last Updated : 19 Sep, 2023
Comments
Improve
Suggest changes
19 Likes
Like
Report

Given an array of n elements and an integer k. The task is to find the count of the subarray which has a maximum element greater than K.

Examples : 

Input : arr[] = {1, 2, 3} and k = 2.
Output : 3
All the possible subarrays of arr[] are
{ 1 }, { 2 }, { 3 }, { 1, 2 }, { 2, 3 },
{ 1, 2, 3 }.
Their maximum elements are 1, 2, 3, 2, 3, 3.
There are only 3 maximum elements > 2.

Approach 1: Counting Subarrays having max element <= K and then subtracting from total subarrays.

The idea is to approach problem by counting subarrays whose maximum element is less than or equal to k as counting such subarrays is easier. To find the number of subarray whose maximum element is less than or equal to k, remove all the element which is greater than K and find the number of subarray with the left elements. 

Once we find above count, we can subtract it from n*(n+1)/2 to get our required result. Observe, there can be n*(n+1)/2 possible number of subarray of any array of size n. So, finding the number of subarray whose maximum element is less than or equal to K and subtracting it from n*(n+1)/2 gets us the answer.

Below is the implementation of this approach:

C++
Java Python3 C# JavaScript PHP

Output
3


Time Complexity: O(n).
Auxiliary Space: O(1)

Approach 2: Counting Subarrays having max element > K

In this approach we just simply find the count of subarrays that can be formed by including an element at index i which is greater than K. Therefore, if suppose arr [ i ] > K then all the subarrays in which this element is present will have a value which is greater than k, so we just calculate all of these subarrays for every element that is greater than K and add them in answer. We first initialize two variables ans = 0 this contains answer and prev = -1 this keeps track of index of previous element that was greater than K.

To do this we just need three values for every arr [ i ] > K .

  1. Number of subarrays starting from the index i. This will be ( N - i ) . NOTE: In this we have included the subarray containing single element that is this element itself. { arr [ i ] }
  2. Number of subarrays ending at this index i but starting index of these subarrays is after the index prev of previous element that was greater than K, why do we do this? Because for that elements we must have already calculated our answer so we dont want to count same subarrays more than once. So, this value will be comes to be ( i - prev - 1 ) . NOTE: In this we subtract 1 because we have already counted a subarray { arr [ i ] } having itself as single element. See above point note. 
  3. Number of subarrays having starting index less than i but greater than prev , and ending index greater than i. Therefore all subarrays in which arr[i] is in between. This we can calculate by multiplying above two values. Lets say them as L = ( N - i - 1 ) and R = (  i  - prev -1 ). Now we just multiply these L and R because for every 1 index on left side of i there are R index that can make different subarrays basic maths thing. So, this becomes L * R . Notice here in val of L we have actually subtracted 1 if we dont do this then we include index i in our L*R which will mean we have included number 1 type subarrays again. See point 1.   

Below is the implementation of this approach:

C++
Java Python3 C# JavaScript

Output
12


Time Complexity: O(n).

Approach 3 : Sliding Window Technique.

Algorithm:

1. Initialize a variable ans = 0 , a varialble maxElement = 0 and a variable count = 0 .

2. Iterate through the array, doing the following for each element:

  a. If the current element i.e. arr[ i ] is greater than current maximum , update the maximum i.e. maxElement = arr[ i ] and reset the count to 0.

  b. If the current element is less than or eual to the current maximum, then increment the count.

  c. If maxElement is grteater than k, then add count of subarrays to final answer and update the maxElement to current element.

3. Return Final answer.

Here's the the implementation of Sliding window technique.

C++
C Java Python3 C# JavaScript

Output
9


The Sliding Window Technique is contributed by Vaibhav Saroj .

Time Complexity: O( n ).
Space Complexity: O( 1 ).

Practice here Count of Subarrays .


Similar Reads