Open In App

Uniform Binary Search

Uniform Binary Search is an optimization of Binary Search algorithm when many searches are made on same array or many arrays of same size. In normal binary search, we do arithmetic operations to find the mid points. Here we precompute mid points and fills them in lookup table. The array look-up generally works faster than arithmetic done (addition and shift) to find the mid point. 

Examples: 

Input : array={1, 3, 5, 6, 7, 8, 9}, v=3
Output : Position of 3 in array = 2

Input :array={1, 3, 5, 6, 7, 8, 9}, v=7
Output :Position of 7 in array = 5

The algorithm is very similar to Binary Search algorithm, The only difference is a lookup table is created for an array and the lookup table is used to modify the index of the pointer in the array which makes the search faster . Instead of maintaining lower and upper bound the algorithm maintains an index and the index is modified using the lookup table. 




// C++ implementation of above approach
#include <bits/stdc++.h>
using namespace std;
 
const int MAX_SIZE = 1000;
 
// lookup table
int lookup_table[MAX_SIZE];
 
// create the lookup table
// for an array of length n
void create_table(int n)
{
    // power and count variable
    int pow = 1;
    int co = 0;
    do {
        // multiply by 2
        pow <<= 1;
 
        // initialize the lookup table
        lookup_table[co] = (n + (pow >> 1)) / pow;
    } while (lookup_table[co++] != 0);
}
 
// binary search
int binary(int arr[], int v)
{
    // mid point of the array
    int index = lookup_table[0] - 1;
 
    // count
    int co = 0;
 
    while (lookup_table[co] != 0) {
 
        // if the value is found
        if (v == arr[index])
            return index;
 
        // if value is less than the mid value
        else if (v < arr[index])
            index -= lookup_table[++co];
 
        // if value is greater than the mid value
        else
            index += lookup_table[++co];
    }
  return index;
}
 
// main function
int main()
{
 
    int arr[] = { 1, 3, 5, 6, 7, 8, 9 };
    int n = sizeof(arr) / sizeof(int);
 
    // create the lookup table
    create_table(n);
 
    // print the position of the array
    cout << "Position of 3 in array = "
         << binary(arr, 3) << endl;
 
    return 0;
}




// Java implementation of above approach
class GFG
{
     
    static int MAX_SIZE = 1000;
     
    // lookup table
    static int lookup_table[] = new int[MAX_SIZE];
     
    // create the lookup table
    // for an array of length n
    static void create_table(int n)
    {
        // power and count variable
        int pow = 1;
        int co = 0;
        do
        {
            // multiply by 2
            pow <<= 1;
     
            // initialize the lookup table
            lookup_table[co] = (n + (pow >> 1)) / pow;
        } while (lookup_table[co++] != 0);
    }
     
    // binary search
    static int binary(int arr[], int v)
    {
        // mid point of the array
        int index = lookup_table[0] - 1;
     
        // count
        int co = 0;
     
        while (lookup_table[co] != 0)
        {
     
            // if the value is found
            if (v == arr[index])
                return index;
     
            // if value is less than the mid value
            else if (v < arr[index])
            {
                index -= lookup_table[++co];
   
            }
             
            // if value is greater than the mid value
            else
            {
                index += lookup_table[++co];
                
            }
        }
        return index ;
    }
     
    // Driver code
    public static void main (String[] args)
    {
     
        int arr[] = { 1, 3, 5, 6, 7, 8, 9 };
        int n = arr.length;
     
        // create the lookup table
        create_table(n);
     
        // print the position of the array
        System.out.println( "Position of 3 in array = " +
                                    binary(arr, 3)) ;
     
     
    }
}
 
// This code is contributed by Ryuga




# Python3 implementation of above approach
 
MAX_SIZE = 1000
 
# lookup table
lookup_table = [0] * MAX_SIZE
 
# create the lookup table
# for an array of length n
def create_table(n):
     
    # power and count variable
    pow = 1
    co = 0
    while True:
         
        # multiply by 2
        pow <<= 1
 
        # initialize the lookup table
        lookup_table[co] = (n + (pow >> 1)) // pow
        if lookup_table[co] == 0:
            break
        co += 1
 
# binary search
def binary(arr, v):
     
    # mid point of the array
    index = lookup_table[0] - 1
 
    # count
    co = 0
 
    while lookup_table[co] != 0:
 
        # if the value is found
        if v == arr[index]:
            return index
 
        # if value is less than the mid value
        elif v < arr[index]:
            co += 1
            index -= lookup_table[co]
 
        # if value is greater than the mid value
        else:
            co += 1
            index += lookup_table[co]
 
# main function
arr = [1, 3, 5, 6, 7, 8, 9]
n = len(arr)
 
# create the lookup table
create_table(n)
 
# print the position of the array
print("Position of 3 in array = ", binary(arr, 3))
 
# This code is contributed by divyamohan123




// C# implementation of above approach
using System;
     
class GFG
{
     
    static int MAX_SIZE = 1000;
     
    // lookup table
    static int []lookup_table = new int[MAX_SIZE];
     
    // create the lookup table
    // for an array of length n
    static void create_table(int n)
    {
        // power and count variable
        int pow = 1;
        int co = 0;
        do
        {
            // multiply by 2
            pow <<= 1;
     
            // initialize the lookup table
            lookup_table[co] = (n + (pow >> 1)) / pow;
        } while (lookup_table[co++] != 0);
    }
     
    // binary search
    static int binary(int []arr, int v)
    {
        // mid point of the array
        int index = lookup_table[0] - 1;
     
        // count
        int co = 0;
     
        while (lookup_table[co] != 0)
        {
     
            // if the value is found
            if (v == arr[index])
                return index;
     
            // if value is less than the mid value
            else if (v < arr[index])
            {
                index -= lookup_table[++co];
                return index;
            }
             
            // if value is greater than the mid value
            else
            {
                index += lookup_table[++co];
                return index;
            }
        }
        return index ;
    }
     
    // Driver code
    public static void Main ()
    {
     
        int []arr = { 1, 3, 5, 6, 7, 8, 9 };
        int n = arr.GetLength(0);
     
        // create the lookup table
        create_table(n);
     
        // print the position of the array
    Console.WriteLine( "Position of 3 in array = " +
                                    binary(arr, 3)) ;
     
     
    }
}
 
/* This code contributed by PrinciRaj1992 */




<script>
 
// Javascript implementation of above approach
let MAX_SIZE = 1000;
   
// lookup table
let lookup_table = new Array(MAX_SIZE);
lookup_table.fill(0);
   
// Create the lookup table
// for an array of length n
function create_table(n)
{
     
    // Power and count variable
    let pow = 1;
    let co = 0;
     
    while(true)
    {
         
        // Multiply by 2
        pow <<= 1;
   
        // Initialize the lookup table
        lookup_table[co] = parseInt((n + (pow >> 1)) /
                                          pow, 10);
         
        if (lookup_table[co++] == 0)
        {
            break;
        }
    }
}
   
// Binary search
function binary(arr, v)
{
     
    // mid point of the array
    let index = lookup_table[0] - 1;
   
    // count
    let co = 0;
   
    while (lookup_table[co] != 0)
    {
         
        // If the value is found
        if (v == arr[index])
            return index;
   
        // If value is less than the mid value
        else if (v < arr[index])
        {
            index -= lookup_table[++co];
            return index;
        }
           
        // If value is greater than the mid value
        else
        {
            index += lookup_table[++co];
            return index;
        }
    }
    return index ;
}
 
// Driver code
let arr = [ 1, 3, 5, 6, 7, 8, 9 ];
let n = arr.length;
 
// Create the lookup table
create_table(n);
 
// Print the position of the array
document.write("Position of 3 in array = " +
               binary(arr, 3));
                
// This code is contributed by divyeshrabadiya07
 
</script>

Output: 
Position of 3 in array = 1

 

Time Complexity : O(log n).

 Auxiliary Space Complexity : O(log n) 

References : https://en.wikipedia.org/wiki/Uniform_binary_search
 


Article Tags :