TheBhediyaOfDalalStreet's blog

By TheBhediyaOfDalalStreet, history, 20 months ago, In English

Hello frends, in this blog I teach you how to find k_th element in an array of size n. This blog is going be very advanced so please if you are new to progrmaming stay away.

eg. arr = {1, 3, 2, 9, 5, 6, 11} and k = 2 then programs must return 3

Approach:

Define dp(i, j, k) = number of inversions in the array formed by concatenating prefix arr[0 .. i] and arr[j .. n-1] such that the smaller number in the inversion pair is >= k (0 <= i < j < n)

The transitions is trivial, so I am skipping those.

then the answer can be found by returning the value of arr[k — 1]

| Write comment?
»
20 months ago, # |
Rev. 2   Vote: I like it 0 Vote: I do not like it

sir you can just use quick select, it can be reduced to $$$O(n)$$$ and a variant of it (introselect) exists in C++ STL as nth_element

  • »
    »
    20 months ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    No madam, quickselect is very slow algrothm O(N^2) worst case. my algrothm works in O(1) worst case.

    • »
      »
      »
      20 months ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      no sir if you really wanted queries on selecting kth elements you really should've better sorted the array

    • »
      »
      »
      3 months ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      madam 😭