BINARY SEARCH ALGORITHM

PRO ALI SAIVI

PH # +92 317 00 11 200
[email protected]
 BINARY SEARCH ALGORITHM

• BINARY SEARCH ALGORITHM FOLLOW DIVIDE AND CONQUER APPROACH.

• TO PERFORM BINARY SEARCH ARRAY LIST MUST BE IN SORTED ORDER(PREREQUISITE FOR BS).
• PERFORMING A BINARY SEARCH WE NEED TO INDEX POINTERS ONE IS LOW (STARTING INDEX)
AND SECOND IS HIGH (ENDING INDEX).

• A KEY VALUE FOR SEARCH

• A MID POINT VALUE FOR COMPARING KEY VALUE. FORMULA MIDPOINT = (LOW + HIGH)/2
• IF KEY VALUE GREATER THAN MID VALUE THAN HIGH REMAIN SAME & LOW = OLD_MID + 1
• IF KEY VALUE LESS THAN MID VALUE THAN LOW REMAIN SAME & HIGHT = OLD_MID - 1

SaiviCodes © 2020 6/26/2020 2

 Binary Search

3 6 8 12 14 17 25 29 31 36 42 47 53 55 62
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Low Mid High

We Find Key = 42 in above Sorted Array List

Low = 1 We Noticed that the key

High = 15 element is greater than Key = 42
its Midpoint we have to Midpoint = 29
Midpoint = [Low + High]/2 leave the first half and
moved to the second 42 >= 29
Midpoint = [1 + 15]/2 Half.
Midpoint = [16]/2
Midpoint = 8

Do Repeat this process Again & Again until value reach’s. we divide and divide array list
until value reach’s. 3
SaiviCodes © 2020 6/26/2020
 Binary Search

3 6 8 12 14 17 25 29 31 36 42 47 53 55 62
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Mid Low High

We Find Key = 42 in above Sorted Array List and Value of Midpoints less Than Key so Change
the Low Value
Old_Mid = 8
Low = Old_Mid+1 = [8 + 1] = 9 We Noticed that the key
High = 15 element is Less than its Key = 42
Midpoint we have to Midpoint = 47
Midpoint = [Low + High]/2 leave the Second half
and moved to the first 42 <= 47

Midpoint = [9 + 15]/2 Half.

Midpoint = [24]/2
Midpoint = 12
Do Repeat this process Again & Again until value reach’s. we divide and divide array list
SaiviCodes © 2020 until value reach’s. 6/26/2020 4
 Binary Search

3 6 8 12 14 17 25 29 31 36 42 47 53 55 62
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Low mid high

We Find Key = 42 in above Sorted Array List and Value of Midpoints Greater Than Key so
Change the High Value
Old_Mid = 12
We Noticed that the key
Low = 9
element is greater than Key = 42
High = Old_Mid – 1 = [12-1] = 11 Midpoint = 36
its Midpoint we have to
leave the first half and
Midpoint = [Low + High]/2 42 >= 36
moved to the second
Half.
Midpoint = [9 + 11]/2
Midpoint = [20]/2
Midpoint = 10
Do Repeat this process Again & Again until value reach’s. we divide and divide array list
SaiviCodes © 2020 until value reach’s. 6/26/2020 5
 Binary Search

3 6 8 12 14 17 25 29 31 36 42 47 53 55 62
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

mid Low
high
mid
We Find Key = 42 in above Sorted Array List and Value of Midpoints less Than Key so Change
the Low Value
Old_Mid = 10
Low = Old_Mid + 1 = 10 + 1 = 11 Key = 42
High = 11 We Noticed that the key Midpoint = 36
element is Equal than its
Midpoint = [Low + High]/2 Midpoint. Now the 42 == 42
Search is Complete…
Midpoint = [11 + 11]/2
Midpoint = [22]/2
Midpoint = 11
Value Matched we required 42 Value in Array in this Steps Both the Value are Equal and
SaiviCodes © 2020
Binary Search Algorithm Stop… 6/26/2020 6
 SUMMARY

Mid = [Low + High]

No of Steps Low High
/2
1 1 15 [1 + 15]/2 = 8
2 9 15 [9 + 15]/2 = 12
3 9 11 [9 + 11]/2 = 10
4 11 11 [11 + 11]/2 = 11

If we search this value using Linear Search we noticed into array the value = 42 founded at
index = 11 , so we have to search out this key using 11 Comparisons its too hard and take a
lot of Time.
SaiviCodes © 2020 6/26/2020 7
 Binary Search

3 6 8 12 14 17 25 29 31 36 42 47 53 55 62
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Low Mid High

We Find Key = 30 in above Sorted Array List

Low = 1 We Noticed that the key

High = 15 element is greater than Key = 30
its Midpoint we have to Midpoint = 29
Midpoint = [Low + High]/2 leave the first half and
moved to the second 30 >= 29
Midpoint = [1 + 15]/2 Half.
Midpoint = [16]/2
Midpoint = 8

Do Repeat this process Again & Again until value reach’s. we divide and divide array list
SaiviCodes © 2020
until value reach’s. 6/26/2020 8
 Binary Search

3 6 8 12 14 17 25 29 31 36 42 47 53 55 62
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Mid Low High

We Find Key = 30 in above Sorted Array List and Value of Midpoints less Than Key so Change
the Low Value
Old_Mid = 8
Low = Old_Mid+1 = [8 + 1] = 9 We Noticed that the key
High = 15 element is Less than its Key = 30
Midpoint we have to Midpoint = 47
Midpoint = [Low + High]/2 leave the Second half
and moved to the first 30 <= 47

Midpoint = [9 + 15]/2 Half.

Midpoint = [24]/2
Midpoint = 12
Do Repeat this process Again & Again until value reach’s. we divide and divide array list
SaiviCodes © 2020
until value reach’s. 6/26/2020 9
 Binary Search

3 6 8 12 14 17 25 29 31 36 42 47 53 55 62
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Low high Mid

We Find Key = 30 in above Sorted Array List and Value of Midpoints greater Than Key so
Change the High Value
Old_Mid = 12
We Noticed that the key
Low = 9
element is greater than Key = 30
High = Old_Mid – 1 = [12-1] = 11 Midpoint = 36
its Midpoint we have to
leave the first half and
Midpoint = [Low + High]/2 30 <= 36
moved to the second
Half.
Midpoint = [9 + 11]/2
Midpoint = [20]/2
Midpoint = 10
Do Repeat this process Again & Again until value reach’s. we divide and divide array list
SaiviCodes © 2020 until value reach’s. 6/26/2020 10
 Binary Search

3 6 8 12 14 17 25 29 31 36 42 47 53 55 62
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Low
high
mid

We Find Key = 42 in above Sorted Array List and Value of Midpoints greater Than Key so
Change the High Value
Old_Mid = 10
Low = 9 Key = 42
High = Old_Mid – 1 = [10-1] = 9 We Noticed that the key Midpoint = 36
element is Equal than its
Midpoint = [Low + High]/2 Midpoint. Now the 30 <= 31
Search is Complete…
Midpoint = [9 + 9]/2
Midpoint = [18]/2
Midpoint = 9
Do Repeat this process Again & Again until value reach’s. we divide and divide array list
until value reach’s. 6/26/2020 11
SaiviCodes © 2020
 Binary Search

3 6 8 12 14 17 25 29 31 36 42 47 53 55 62
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Low
high
mid
We Find Key = 42 in above Sorted Array List and Value of Midpoints greater Than Key so
Change the High Value

We Noticed that the High

Old_Mid = 9
Value is Less than Low
Low = 9
Value now comes to the
High = Old_Mid – 1 = [9-1] = 8
point that High value
never ever be smallest
Low > High (Wrong Case)
value over the Low
Value

Algorithm Stop due to Value not Found in Array

SaiviCodes © 2020 6/26/2020 12
 SUMMARY

Mid = [Low + High]

No of Steps Low High
/2
1 1 15 [1 + 15]/2 = 8
2 9 15 [9 + 15]/2 = 12
3 9 11 [9 + 11]/2 = 10
4 9 9 [9 + 9]/2 = 9
5 9 8 Stop Error Occurs

SaiviCodes © 2020 6/26/2020 13

 ALGORITHM

SaiviCodes © 2020 6/26/2020 14

 Binary Search

3 6 8 12 14 17 25 29 31 36 42 47 53 55 62
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Log n 8

4 12

2 6 10 14

1 3 5 7 9 11 13 15

15+1 = 16 => log16 base 2 = 4

SaiviCodes © 2020 4 Comparison are Required 6/26/2020 15
 Binary Search

3 6 8 12 14 17 25 29 31 36 42 47 53 55 62
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Log n 8 Min Time – 1 O(1)

Max Time – log n O(log n)
Best Case: O(1)
4 12 Worst Case: O(log n)

2 6 10 14

1 3 5 7 9 11 13 15

If we search Value 65 or 57 so either these values are 56 65

not in array list so it will take 4 comparison as Well 6/26/2020
16
 SUMMARY

• BEST CASE IT IS IF YOU ARE SEARCHING FOR A KEY ELEMENT IN THE MIDDLE OF
ARRAY LIST THAT IS BEST CASE AND TIME IS 1 O(1).

• WORST CASE IF YOU ARE SEARCHING FOR A KEY ELEMENT PRESENT ON THE
LEAFE AND AT THE TIME IS LOG N IN WORST CASE O(LOG N).

SaiviCodes © 2020 6/26/2020 17

 Binary Search

3 6 8 12 14 17 25 29 31 36 42 47 53 55 62
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Algorithm RBinarySearch(L, H, Key) T(n)

if (L == H)
if (A[L] == key)
return L; 1 1 n=1
else
return 0; T(n) =
else
𝒏
Mid = (L + H)/2; 1 T(𝟐) + 1 n>1
if (Key == A[Mid]) 1
return Mid;
if (Key < A[Mid]) 1 T(n) = 1 T( ) + 1
𝒏
return RBinarySearch(L, Mid-1,Key); 𝟐
𝒏 log 𝟐 𝟏 = 0
else T( )
𝟐 𝒏𝟎 O (log 𝒏)
return RBinarySearch(Mid+1,H,Key);
SaiviCodes © 2020 6/26/2020 18