Algorithms notes 2 tutorials duniya

Download FREE Computer Science Notes, Programs,
Projects, Books for any university student of BCA,
MCA, B.Sc, M.Sc, B.Tech CSE, M.Tech at
https://www.tutorialsduniya.com
Please Share these Notes with your Friends as well
TUTORIALSDUNIYA.COM
Computer Science Notes
Design and Analysis of
Algorithms Notes

INTRODUCTION TO ALGORITHMS UNIT-1
1. Introduction:
Algorithm: The word algorithm came from the name of a Persian mathematician Abu Jafar
Mohammed Ibn Musa Al Khowarizmi (ninth century). An algorithm is simply s set of rules used
to perform some calculations either by hand or more usually on a machine (computer).
Definition: An algorithm is a finite set of instructions that accomplishes a particular task.
Another definition is a sequence of unambiguous instructions for solving a problem i.e, for
obtaining a required output for any legitimate (genuine) input in a finite amount of time.
In addition all algorithms must satisfy the following criteria (characteristics).
1. Input: zero or more quantities are externally supplied as input.
Consider Fibonacci numbers program, here aim of the problem is to display ten
Fibonacci numbers. No input is required; in the problem itself this is clearly mentioned as ten
Fibonacci values. So zero items required for input.
Another problem is displaying given numbers of evens, so user should accept how
many evens required. Based on the user input the number of evens is to be displayed. So, one data
item is required as input.
2. Output: At least one quantity is produced by given algorithm as output.
In the case of Fibonacci numbers program after executing the program, first ten
Fibonacci values displayed as output.
In second case, based on user input it should display given number of evens. An input of
negative number is wrong, should display proper error message as output. So this program
displays at least one output as error message, or number if outputs that show given number of
steps.
3. Definiteness: Each instruction is clear and unambiguous i.e. each step must be easy to
understand and convey only a single meaning.
4. Effectiveness: each instruction must be very basic, so that it can be carried out by a
person using only pencil and paper.
This step is common in both Fibonacci and primes. For example, if user enters a negative
numbers as input in evens, if you have a step like
Step: If N < 0 then
Go to ERROR
A wrong instruction given as go to ERROR, those kinds of instructions should not be there
in an algorithm.
5. Finiteness: If we can trace out the instructions of an algorithm then for all cases, the
algorithm terminate after a finite number of steps.
Either in the case of Fibonacci or even numbers problem should be solved in some number
of steps. For example, continuous display or Fibonacci series without termination leads to
abnormal termination.
Criteria for Algorithms
Input: Zero or more inputs
Output: At least one output.
Finiteness: N number of steps.
Definiteness: Clear algorithm step.
Effectiveness: A carried out step.
Page 1 of 13
TutorialsD
uniya.com
Visit https://www.tutorialsduniya.com for Notes, books, programs, question papers with solutions etc.

2. Process for Design and analysis of algorithms:
Understand the problem
Solution as an algorithm
Yes
No
No
Yes
Prove Correctness
Analyse the algorithm
is it efficient
Code the algorithm
Algorithm Design
technique
Fig: Process for design and analysis of algorithms
1. Understand the problem: This is very crucial phase. If we did any mistake in this step
the entire algorithm becomes wrong. So before designing an algorithm is to understand the
problem first.
2. Solution as an algorithm (Exact vs approximation solving): Solve the problem exactly if
possible. Even though some problems are solvable by exact method, but they are not faster when
compared to approximation method. So in that situation we will use approximation method.
3. Algorithm techniques: In this we will use different design techniques like,
i) Divide-and-conquer
ii) Greedy method
iii) Dynamic programming
iv) Backtracking
v) Branch and bound…. etc.,
4. Prove correctness: once algorithm has been specified, next we have to prove its
correctness. Usually testing is used for proving correctness.
5. Analyze an algorithm: Analyzing an algorithm means studying the algorithm behavior
ie., calculating the time complexity and space complexity. If the time complexity of algorithm is
more then we will use one more designing technique such that time complexity should be
minimum.
6. Coding an algorithm: after completion of all phases successfully then we will code an
algorithm. Coding should not depend on any program language. We will use general notation
(pseudo-code) and English language statement. Ultimately algorithms are implemented as
computer programs.
3. Types of Algorithms:
There are four types of algorithms
1. Approximate algorithm.
2. Probabilistic algorithm.
3. Infinite algorithm.
4. Heuristic algorithm.
Page 2 of 13
TutorialsD
uniya.com

1. Approximate Algorithm: An algorithm is said to approximate if it is infinite and repeating.
1.
2
3
= 1.414
= 1.713
= 3.14
Ex:
etc...
2. Probabilistic algorithm: If the solution of a problem is uncertain then it is called as
probabilistic algorithm. Ex: Tossing of a coin.
3. Infinite Algorithm: An algorithm which is not finite is called as infinite algorithm.
Ex: A complete solution of a chessboard, division by zero.
4. Heuristic algorithm: Giving fewer inputs getting more outputs is called the Heuristic
algorithms.
Ex: All Business Applications.
4. Criteria’s (or) Issues for algorithm:
There are various issues in the study of algorithms;
1. How to devise algorithms: The creation of an algorithm is a logical activity which may
never be fully automated.
2. How to express algorithms: We shall express all of our algorithms using the best
principles of structuring.
3. How to validate algorithms: After creation of algorithms is to validate algorithms. The
process of checking an algorithm computes the correct answer for all possible legal inputs is called
algorithm validation. The purpose of validation of algorithm is to find whether algorithm works
properly without being dependent upon programming languages.
4. How to analyze algorithms: Analysis of algorithm is a task of determining how much
computing time and storage is required by an algorithm. Analysis of algorithms is also called
performance analysis. The behavior of algorithm in best case, worst case and average case needs
to be obtained.
5. How to test a program: Testing a program really consists of two phases:
i). Debugging: While debugging a program, it is checked whether program produces faulty
results for valid set of input and if it is found then the program has to be corrected.
ii). Profiling or performance measuring: Profiling is a process of measuring time and space
required by a corrected program for valid set of inputs.
5 Specification of algorithm:
There are various ways by which we can specify an algorithm.
Algorithm
Using natural language
Pseudocode
Flow chart
Program (Using programming language)
It is very easy to specify an algorithm using natural language. But many times
specification of algorithm by using natural language is not clear, and may require brief
description.
Example: Write an algorithm to perform addition of two numbers.
Step 1: Read the first number, say ‘a’.
Step 2: Read the second number, say ‘b’.
Step 3: Add the two numbers and store the result in a variable ‘c’.
Step 4: Display the result.
Page 3 of 13
TutorialsD
uniya.com

Such a specification creates difficulty, while actually implementing it (difficulty in
converting into source code). Hence many programmers prefer to have specification of algorithm
by means of pseudo-code.
Another way of representing the algorithm is by flow chart. Flow chart is a graphical
representation of an algorithm, but flowchart method work well only if the algorithm is small
and simple.
Pseudo-Code for expressing Algorithms
Based on algorithm there are two more representations used by programmer and those are
flow chart and pseudo-code. Flowchart is a graphical representation of an algorithm. Similarly
pseudo-code is a representation of algorithm in which instruction sequence can be given with the
help of programming constructs. It is not a programming language since no pseudo language
compiler exists.
The general procedure for writing the pseudo-code is presented below-
1. Comments begin with // and continue until the end of line
2. A block of statements (compound statement) are represented using { and } for example if
statement, while loop, functions etc.,.
Example
{
Statement 1;
Statement 2;
.........
.........
}
3. The delimiters [;] are used at the end of the each statement.
4. An identifier begins with a letter. Example: sum, sum5, a; but not in 5sum, 4a etc.,.
5. Assignment of values to the variables is done using the assignment operators as := or .
6. There are two Boolean values TRUE and FALSE.
Logical operators: AND, OR, NOT.
Relational operators: <, , ≥,,=,.
Arithmetic operators: +, -, *, /, %;
7. The conditional statement if-then or if-then-else is written in the following form.
If (condition) then (statement)
If (condition) then (statement-1) else (statement-2)
‘If’ is a powerful statement used to make decisions based as a condition. If a condition is
true the particular block of statements are execute.
Example
if(a>b) then
{
write("a is big");
}
else
{
write("b is big");
}
Page 4 of 13
TutorialsD
uniya.com

8. Case statement
case
{
:(condition -1): (statement-1)
:(condition -2): (statement-2)
:(condition -n): (statement-n)
..............
..............
else :(statement n+1);
}
If condition -1 is true, statement -1 executed and the case statement is exited. If
statement -1 is false, condition -2 is evaluated. If condition -2 is true, statement-2 executed
and so on. If none of the conditions are true, statement –(n+1) is executed and the case
statement is exited. The else clause is optional.
9. Loop statements:
For loop:
i). The general form of the for loop is
for variable:=value 1 to value n step do
{
Statement -1;
Statement -1;
.......
.......
Statement -n;
}
Example:
for i:=1 to 10 do
{
write(i); //displaying numbers from 1 to 10
i:=i+1;
}
ii). While loop:
The general form of the while loop is
while <condition> do
{
<statement 1>
<statement 2>
........
........
<statement n>
}
Example:
i:=1;
while(i<=10)do
{
write (i);//displaying numbers from 1 to 10
i:=1+1;
}
Note that the statements of while loop are executed as long as <condition> is true.
iii). Repeat-until loop:
The general form of repeat-until is-
repeat
{
<statement 1>
<statement 2>
......
......
<statement n>
}until <condition>
Example
i:=1;
repeat
{
write (i);
i:=i+1;
}
until (i>10);
Note that the statements are executed as long as <condition> is false.
Page 5 of 13
TutorialsD
uniya.com

10. Break: this statement is exit from the loop.
11. Elements of array are accessed using [ ].
For example, if A is an one-dimensional array, then ith
element can be accessed using
A[i]. If A is two-dimensional array, then (i, j)th
element can be accessed using A[i,j].
12. Procedures (functions): There is only one type of procedure:
An algorithm consists of a heading and a body.
Algorithm Name (<parameter list>)
{
body of the procedure
}
procedure name of the procedure
Syntax:
13.Compound data-types can be formed with records
Syntax:
Name = record
{
data-type -1 data 1;
data-type -2 data 2;
data-type -n data n;
}
Example
Employee =record
{
int no;
char name[10];
float salary;
}
Example 1: Write an algorithm to find the sum of n numbers.
Algorithm sum(n)
{
total:=0;
for i:=1 to n do
total:= total + i;
i:=i+1;
}
Example 2: Write an algorithm to perform Multiplication of two matrices.
Algorithm Multiplication (A, B, n)
{
for i:=1 to n do
for j:=1 to n do
C[i,j]:=0;
for k:=1 to n do
C[i,j]:=C[i,j]+A[i,k]*B[k,j];
}
6. Performance Analysis:
Performance analysis or analysis of algorithms refers to the task of determining the
efficiency of an algorithm i.,e how much computing time and storage an algorithm requires to
run (or execute). This analysis of algorithm helps in judging the value of one algorithm over
another.
To judge an algorithm, particularly two things are taken into consideration
1. Space complexity
2. Time complexity.
Space Complexity: The space complexity of an algorithm (program) is the amount of
memory it needs to run to completion. The space needed by an algorithm has the following
components.
Page 6 of 13
TutorialsD
uniya.com

1. Instruction Space.
2. Data Space.
3. Environment Stack Space.
Instruction Space: Instruction space is the space needed to store the compiled version of the
program instructions. The amount of instruction space that is needed depends on factors such as-
i). The compiler used to compile the program into machine code.
ii). The compiler options in effect at the time of compilation.
iii). The target computer, i.,e computer on which the algorithm run.
Note that, one compiler may produce less code as compared to another compiler, when the
same program is compiled by these two.
Data Space: Data space is the space needed to store all constant and variable values. Data
space has two components.
i). Space needed by constants, for example 0, 1, 2.134.
ii). Space needed by dynamically allocated objects such as arrays, structures, classes.
Environmental Stack Space: Environmental stack space is used during execution of
functions. Each time function is involved the following data are saved as the environmental
stack.
i). The return address.
ii). Value of local variables.
iii). Value of formal parameters in the function being invoked.
Environmental stack space is mainly used in recursive functions. Thus, the space requirement
of any program p may therefore be written as
Space complexity S(P) = C + Sp (Instance characteristics).
This equation shows that the total space needed by a program is divided into two parts.
 Fixed space requirements(C) is independent of instance characteristics of the inputs and
outputs.
- Instruction space
- Space for simple variables, fixed-size structure variables, constants.
 A variable space requirements (SP(1)) dependent on instance characteristics 1.
- This part includes dynamically allocated space and the recursion stack space.
Example of instance character is:
Examples: 1
Algorithm NEC (float x, float y, float z)
{
Return (X + Y +Y * Z + (X + Y +Z)) /(X+ Y) + 4.0;
}
In the above algorithm, there are no instance characteristics and the space needed by X, Y, Z is
independent of instance characteristics, therefore we can write,
S(XYZ) =3+0=3
One space each for X, Y and Z
Space complexity is O(1).
Examples: 2
Algorithm ADD ( float [], int n)
{
sum = 0.0;
for i=1 to n do
sum=sum+X[i];
return sum; }
Page 7 of 13
TutorialsD
uniya.com

Here, atleast n words since X must be large enough to hold the n elements to be summed.
Here the problem instances is characterized by n, the number of elements to be summed. So, we
can write,
S(ADD) =3+n
3-one each for n, I and sum
Where n- is for array X[],
Space complexity is O(n).
Time Complexity
The time complexity of an algorithm is the amount of compile time it needs to run to
completion. We can measure time complexity of an algorithm in two approaches
1. Priori analysis or compile time
2. Posteriori analysis or run (execution) time.
In priori analysis before the algorithm is executed we will analyze the behavior of the
algorithm. A priori analysis concentrates on determining the order if execution of statements.
In Posteriori analysis while the algorithm is executed we measure the execution time.
Posteriori analysis gives accurate values but it is very costly.
As we know that the compile time does not depend on the size of the input. Hence, we will
confine ourselves to consider only the run-time which depends on the size of the input and this
run-time is denoted by TP(n). Hence
Time complexity T(P) = C + TP(n).
The time (T(P)) taken by a program P is the sum of the compile time and execution time.
The compile time does not depend on the instance characteristics, so we concentrate on the
runtime of a program. This runtime is denoted by tp (instance characteristics).
The following equation determines the number of addition, subtraction, multiplication,
division compares, loads stores and so on, that would be made by the code for p.
tp(n) = CaADD(n)+ CsSUB(n)+ CmMUL(n)+ CdDIV(n)+……………..
where n denotes instance characteristics, and Ca, Cs, Cm, Cd and so on…..
As denote the time needed for an addition, subtraction, multiplication, division and so on,
and ADD, SUB, MUL, DIV and so on, are functions whose values are the number of additions,
subtractions, multiplications, divisions and so on. But this method is an impossible task to find
out time complexity.
Another method is step count. By using step count, we can determine the number if steps
needed by a program to solve a particular problem in 2 ways.
Method 1: introduce a global variable “count”, which is initialized to zero. So each time a
statement in the signal program is executed, count is incremented by the step count of that
statement.
Page 8 of 13
TutorialsD
uniya.com

Example:
Algorithm Sum(a, n)
{
s:=0;
for i:=1 to n do
{
s:=s+a[i];
}
return s;
}
Algorithm sum with count statement added
count:=0;
Algorithm Sum(a,n)
{
s:=0;
count:=count+1;
for i:=1 to n do
{
count:=count +1;
s:=s+a[i];
count:=count+1;
}
count:=count+1; //for last time of for loop
count:=count+1; //for return statement
return s;
}
Thus the total number of steps are 2n+3
Method 2: The second method to determine the step count of an algorithm is to build a
table in which we list the total number of steps contributed by each statement.
1. Algorithm Sum(a, n)
2. {
3. s:=0;
4. for i:=1 to n do
5. s:=s+a[i];
6. return s;
7. }
Total
Statement S/e Frequency Total steps
2n+3 steps
0
0
1
1
1
1
0
-
-
1
n+1
n
1
-
0
0
1
n+1
n
1
0
Ex:
The S/e (steps per execution) of a statement is the amount by which the count changes as
a result of the execution of that statement. The frequency determines the total number of times
each statement is executed.
Complexity of Algorithms:
1. Best Case: Inputs are provided in such a way that the minimum time is required to
process them.
2. Average Case: The amount of time the algorithm takes on an average set of inputs.
3. Worst Case: The amount of time the algorithm takes on the worst possible set of inputs.
3 4 5 6 7 9 10 12 15
Linear Search
A 1 2 3 4 5 6 7 8 9
Example:
Best Case: If we want to search an element 3, whether it is present in the array or not. First, A(1)
is compared with 3, match occurs. So the number of comparisons is only one. It is observed that
search takes minimum number of comparisons, so it comes under best case.
Time complexity is O(1).
Page 9 of 13
TutorialsD
uniya.com

Average Case: If we want to search an element 7, whether it is present in the array or not.
First, A(1) is compared with 7 i,.e, (3=7), no match occurs. Next, compare A(2) and 7, no match
occurs. Compare A(3) and A(4) with 7, no match occurs. Up to now 4 comparisons takes place.
Now compare A(5) and 7 (i.,e, 7=7), so match occurs. The number of comparisons is 5. It is
observed that search takes average number of comparisons. So it comes under average case.
Note: If there are n elements, then we require n/2 comparisons.
. .
.
Time complexity is O
n
2
= O(n) (we neglect constant)
Worst Case: If we want to search an element 15, whether it is present in the array or not.
First, A(1) is compared with 15 (i.,e, 3=15), no match occurs. Continue this process until either
element is found or the list is exhausted. The element is found at 9th
comparison. So number of
comparisons are 9.
Time complexity is O(n).
Note: If the element is not found in array, then we have to search entire array, so it comes under
worst case.
7. Asymptotic Notation:
Accurate measurement of time complexity is possible with asymptotic notation.
Asymptotic complexity gives an idea of how rapidly the space requirement or time requirement
grow as problem size increase. When there is a computing device that can execute 1000 complex
operations per second. The size of the problem is that can be solved in a second or minute or an
hour by algorithms of different asymptotic complexity. In general asymptotic complexity is a
measure of algorithm not problem. Usually the complexity of an algorithm is as a function
relating the input length to the number of steps (time complexity) or storage location (space
complexity). For example, the running time is expressed as a function of the input size ‘n’ as
follows.
f(n)=n4
+100n2
+10n
+50 (running time)
There are four important asymptotic notations.
1. Big oh notation (O)
2. Omega notation ().
3. Theta notation ()
Let f(n) and g(n) are two non-negative functions.
Big oh notation
Big oh notation is denoted by ‘O’. it is used to describe the efficiency of an algorithm. It is
used to represent the upper bound of an algorithms running time. Using Big O notation, we can
give largest amount of time taken by the algorithm to complete.
Definition: Let f(n) and g(n) be the two non-negative functions. We say that f(n) is said to be
O(g(n)) if and only if there exists a positive constant ‘c’ and ‘n0‘ such that,
f(n)c*g(n) for all non-negative values of n, where n≥n0.
Here, g(n) is the upper bound for f(n).
Page 10 of 13
TutorialsD
uniya.com

Ex: Let f(n) = 2n
4
+ 5n2 + 2n +3
2n
4
+ 5n
4
+ 2n
4
+3n
4
(2+5+2+3)n
4
12n
4
.
f(n)=12n
4
This implies g(n)=n
4
, n
c=12 and n =1
f(n)=O(n
4
). .
.
. .
.
. .
.
<
<
<
>1
0
n0
f(n)
c*g(n)
The above definition states that the function ‘f’ is almost ‘c’ times the function ‘g’ when ‘n’ is
greater than or equal to n0.
This notion provides an upper bound for the function ‘f’ i.,e, the function g(n) is an upper
bound on the value of f(n) for all n, where n≥ n0.
Big omega notation
Big omega notation is denoted by ‘’. It is used to represent the lower bound of an
algorithms running time. Using big omega notation we can give shortest amount of time taken by
the algorithm to complete.
Definition: The function f(n)= (g(n)) (read as for of n is omega of g of n) if and only if
there exist positive constants ‘c’ and ‘n0’ such that,
f(n) ≥ c*g(n) for all n, n≥n0
Example:
. .
.
>1
g(n)=n
4
, c=2 and n =1
f(n)=. .
.
. .
.
0
(n
4
)
f(n) 2n
4, n>
c*g(n)
n0
f(n)
Let f(n) = 2n
4
+ 5n2 + 2n +3
2n
4 (for example as n
lower order oterms
are insignificant)
> ,
Big Theta notation
The big theta notation is denoted by ‘’. It is in between the upper bound and lower
bound of an algorithms running time.
Definition: Let f(n) and g(n) be the two non-negetive functions. We say that f(n) is said to
be (g(n)) if and only if there exists a positive constants ‘c1’ and ‘c2’, such that,
c1g(n)  f(n)  c2g((n) for all non-negative values n, where n ≥ n0.
The above definition states that the function f(n) lies between ‘c1’times the function g(n)
and ‘c2’, times the function g(n) where ‘c1’ and ‘c2’ are positive constants.
This notation provides both lower and upper bounds for the function f(n) i.,e, g(n) is both
lower and upper bounds on the value of f(n), for large n. in other words theta notation says that
f(n) is both O(g(n)) and (g(n)) for all n, where n≥n0.
Page 11 of 13
TutorialsD
uniya.com

This function f(n) = (g(n)) iff g(n) is both upper and lower bound an f(n).
Example:
n0
f(n)
2n4 2n4 + 5n2 + 2n +3 12n4
2n4 f(n) 12n4 , n 1
g(n) = n4
c1=2, c2=12 and n0=1
f(n)=(n4)
. .
.
. .
.
. .
.
f(n) = 2n4 + 5n2 + 2n +3
c1*g(n)
c2*g(n)
n
Little ‘oh’ notation
Little oh notation is denoted by “o”. the asymptotic upper bound provided by O-notation
may or may not be asymptotically tight. The bound 2n2
=O(n2
) is asymptotically tight, but the
bound 2n=O(n2
) is not. We use o-notation to denote an upper bound that is not asymptotically
tight.
Definition: f(n)=o(g(n)), iff f(n)<c.g(n) for any positive constants c>0, n0>0 and n> n0.
Or
lim
f(n)
g(n)
=0
n
1. Algorithm Sum(a, n)
2. {
3. s:=0;
4. for i:=1 to n do
5. s:=s+a[i];
6. return s;
7. }
Total
Statement S/e Frequency Total steps
0
0
1
1
1
1
0
-
-
1
n+1
n
1
-
Ex:
(0)
(0)
(1)
(n)
(n)
(1)
(0)
(n)
The aymptotic complexity of sum is as follows
Find the time complexity for sum f given array elememts
The time complexity can be calculated as follows-
First find out the basic operation of the above algorithm is-
S:=S+ a[i] i.e., addition in the loop is the basic operation.
The basic operation is executed every time the loop is executed.
Page 12 of 13
TutorialsD
uniya.com

TutorialsDuniya.com
Download FREE Computer Science Notes, Programs, Projects,
Books PDF for any university student of BCA, MCA, B.Sc,
B.Tech CSE, M.Sc, M.Tech at https://www.tutorialsduniya.com
 PHP Notes
 Python Programming
 R Programming
 Software Engineering
 System Programming
 Theory of Computation
 Unix Network Programming
 Web Design & Development
 Information Security
 Internet Technologies
 Java Programming
 JavaScript & jQuery
 Machine Learning
 Microprocessor
 Operating System
 DBMS & SQL Notes
 Data Analysis & Visualization
 Data Mining
 Data Science
 Data Structures
 Deep Learning
 Digital Image Processing
 Discrete Mathematics
 Computer System Architecture  Operational Research
 Algorithms Notes
 Artificial Intelligence
 Android Programming
 C & C++ Programming
 Combinatorial Optimization
 Computer Graphics
 Computer Networks

T(n) =
i=1
n
1Thus time complexity
= 1 + 1 + 1 + 1 n
= n
T(n) = O(n). .
.
8 Probabilistic Analysis:
Probabilistic analysis of algorithms is an approach to estimate the complexity of an
algorithm. It uses the probability in the analysis of problems. It starts from an assumption about a
probabilistic distribution of the set of all possible inputs. This assumption is then used to design an
efficient algorithm or to compute an expected running time of a known algorithm.
The following is the simple example as probabilistic average case analysis.
Example: Consider linear search algorithm which searches a target element say x, in the
given list of size n. in the worst case, the algorithm will examine all n elements in the list before
terminating.
For a probabilistic average-case analysis, it is generally assumed that all possible
terminations are equally likely-that is, the probability that x, will be found at position 1 is 1/x at
position 2 is 1/n and so on.
The average search cost is therefore the sum of all possible search costs each multiplied
by their associated probability.
For example, if n=5, we would have
Average search cost=1/5(1 +2 +3 +4 +5)=3.
In general case we have
Average search cost =1/n(n(n+1)/2)=(n+1)/2
Probabilistic analysis is mainly useful in estimate running time of an algorithm, calculating
search costs in a searching algorithm etc.
9. Amortized Analysis:
Amortized analysis refers to finding the average running time per operation, over a worst
case sequence of operations. That is the main goal of amortized analysis is to analyze the time
per operation for a series of operations. Sometimes single operation might be expensive; in that
case amortized analysis specifies average over a sequence of operations. Amortized cost per
operation for a sequence of n operations is the total cost of operations divided by n.
For example, if we have 100 operations at cost 1, followed by one operation at cost 100,
then amortized cost per operation is 200/101 < 2. Amortized analysis does not allow random
selection of input.
The average case analysis and amortized analysis are different. In average case analysis,
we are averaging over all possible inputs whereas in amortized analysis we are averaging over a
sequence of operations.
Amortized analysis does not allow random selection of input.
There are several techniques used in amortized analysis.
1. Aggregate Analysis: In this type of analysis upper bound T(n) on the total cost of a
sequence of n operations is decided, then the average cost is calculated as T(n)/n.
2. Accounting Method: In this method the individual cost of each operation is determined,
by combining immediate execution time and its influence on the running time of future operations.
3. Potential Method: It is like the accounting method, but overcharges operations early to
compensate for undercharges later.
Page 13 of 13
TutorialsD
uniya.com

UNIT-2 DIVIDE AND CONQUER
Divide-and-conquer method: Divide-and-conquer are probably the best known general
algorithm design technique. The principle behind the Divide-and-conquer algorithm design
technique is that it is easier to solve several smaller instance of a problem than the larger one.
The “divide-and-conquer” technique involves solving a particular problem by dividing it
into one or more cub-problems of smaller size, recursively solving each sub-problem and then
“merging” the solution of sub-problems to produce a solution to the original problem.
Divide-and-conquer algorithms work according to the following general plan.
1. Divide: Divide the problem into a number of smaller sub-problems ideally of about the
same size.
2. Conquer: The smaller sub-problems are solved, typically recursively. If the sub-problem
sizes are small enough, just solve the sub-problems in a straight forward manner.
3. Combine: If necessary, the solution obtained the smaller problems are connected to get
the solution to the original problem.
The following figure shows-
Fig: Divide and Conquer typical case
Control abstraction for divide-and-conquer technique:
Control abstraction means a procedure whose flow of control is clear but whose primary
operations are satisfied by other procedure whose precise meanings are left undefined.
Algorithm DandC(p)
{
if small (p) then
return S(p)
else
{
Divide P into small instances P1, P2, P3……..Pk, k≥1;
Apply DandC to each of these sub-problems;
return combine (DandC(P1), DandC(P1),…. (DandC(Pk);
}
}
Page 1 of 13
TutorialsD
uniya.com

Algorithm: Control abstraction for divide-and-conquer
DandC(p) is the divide-and-conquer algorithm, where P is the problem to be solved.
Small(p) is a Boolean valued function(i.e., either true or false) that determines whether the input
size is small enough that the answer can be computed without splitting. If this, is so the function
S is invoked. Otherwise the problem P is divided into smaller sub-problems. These sub-problems
P1, P2, P3……..Pk, are solved by receive applications of DandC.
Combine is a function that combines the solution of the K sub-problems to get the
solution for original problem ‘P’.
Example: Specify an application that divide-and-conquer cannot be applied.
Solution: Let us consider the problem of computing the sum of n numbers a0, a1,…..an-1. If
n>1, we divide the problem into two instances of the same problem. That is to compute the sum
of the first [n/2] numbers and to compute the sum of the remaining [n/2] numbers. Once each of
these two sum is compute (by applying the same method recursively), we can add their values to
get the sum in question-
a0+ a1+….+an-1= (a0+ a1+….+a[n/2]-1)+ a[n/2]-1+………+ an-1).
For example, the sum of 1 to 10 numbers is as follows-
(1+2+3+4+………………..+10) = (1+2+3+4+5)+(6+7+8+9+10)
= [(1+2) + (3+4+5)] + [(6+7) + (8+9+10)]
= …..
= …..
= (1) + (2) +…………..+ (10).
This is not an efficient way to compute the sum of n numbers using divide-and-conquer
technique. In this type of problem, it is better to use brute-force method.
Applications of Divide-and Conquer: The applications of divide-and-conquer methods are-
1. Binary search.
2. Quick sort
3. Merge sort.
Binary Search:
Binary search is an efficient searching technique that works with only sorted lists. So the
list must be sorted before using the binary search method. Binary search is based on divide-and-
conquer technique.
The process of binary search is as follows:
The method starts with looking at the middle element of the list. If it matches with the key
element, then search is complete. Otherwise, the key element may be in the first half or second
half of the list. If the key element is less than the middle element, then the search continues with
the first half of the list. If the key element is greater than the middle element, then the search
continues with the second half of the list. This process continues until the key element is found
or the search fails indicating that the key is not there in the list.
Consider the list of elements: -4, -1, 0, 5, 10, 18, 32, 33, 98, 147, 154, 198, 250, 500.
Trace the binary search algorithm searching for the element -1.
Page 2 of 13
TutorialsD
uniya.com

Sol: The given list of elements are:
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
-4 -1 0 5 10 18 27 32 33 98 147 154 198 250 500
Low High
Searching key '-1': Here the key to search is '-1'
First calculate mid;
Mid = (low + high)/2
= (0 +14) /2 =7
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
-4 -1 0 5 10 18 27 32 33 98 147 154 198 250 500
Low HighMid
>< First Half Second Half ><
Here, the search key -1 is less than the middle element (32) in the list. So the search process
continues with the first half of the list.
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
-4 -1 0 5 10 18 27 32 33 98 147 154 198 250 500
Low High
Now mid = (0+6)/2
=3.
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
-4 -1 0 5 10 18 27 32 33 98 147 154 198 250 500
Low HighMid
>< First Half Second Half ><
The search key ‘-1’ is less than the middle element (5) in the list. So the search process
continues with the first half of the list.
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
-4 -1 0 5 10 18 27 32 33 98 147 154 198 250 500
Low High
Now mid= ( 0+2)/2
=1
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
-4 -1 0 5 10 18 27 32 33 98 147 154 198 250 500
Low HighMid
Here, the search key -1 is found at position 1.
Page 3 of 13
TutorialsD
uniya.com

The following algorithm gives the iterative binary Search Algorithm
Algorithm BinarySearch(a, n, key)
{
// a is an array of size n elements
// key is the element to be searched
// if key is found in array a, then return j, such that
//key = a[i]
//otherwise return -1.
Low: = 0;
High: = n-1;
While (low  high) do
{
Mid: = (low + high)/2;
If ( key = a[mid]) then
Return mid;
Else if (key < a[mid])
{
High: = mid +1;
}
Else if( key > a[mid])
{
Low: = mid +1;
}
}
The following algorithm gives Recursive Binary Search
Algorithms Binsearch ( a, n, key, low, high)
{
// a is array of size n
// Key is the element to be searched
// if key is found then return j, such that key = a[i].
//otherwise return -1
If ( low  high) then
{
Mid: = (low + high)/2;
If ( key = a[mid]) then
Return mid;
Else if (key < a[mid])
Binsearch ( a, n, key, low, mid-1);
Else if ( key > a[mid])
Binsearch ( a, n, key, mid+1, high);
}
Return -1;
}
Advantages of Binary Search: The main advantage of binary search is that it is faster than
sequential (linear) search. Because it takes fewer comparisons, to determine whether the given
key is in the list, then the linear search method.
Page 4 of 13
TutorialsD
uniya.com

Disadvantages of Binary Search: The disadvantage of binary search is that can be applied to
only a sorted list of elements. The binary search is unsuccessful if the list is unsorted.
Efficiency of Binary Search: To evaluate binary search, count the number of comparisons in
the best case, average case, and worst case.
Best Case: The best case occurs if the middle element happens to be the key element. Then only
one comparison is needed to find it. Thus the efficiency of binary search is O(1).
Ex: Let the given list is: 1, 5, 10, 11, 12.
1 5 10 11 12
Low HighMid
Let key = 10.
Since the key is the middle element and is found at our first attempt.
Worst Case: Assume that in worst case, the key element is not there in the list. So the process of
divides the list in half continues until there is only one item left to check.
Items left to search Comparisons so far
16
8
4
2
1
0
1
2
3
4
For a list of size 16, there are 4 comparisons to reach a list of size one, given that there is one
comparison for each division, and each division splits the list size in half.
In general, if n is the size of the list and c is the number of comparisons, then
C = log2 n
Eficiency in worst case = O(log n). .
.
Average Case: In binary search, the average case efficiency is near to the worst case efficiency.
So the average case efficiency will be taken as O(log n).
Efficiency in average case = O (log n).
Binary Search
Best Case O(1)
Average Case O( log n)
Worst Case O(log n)
Space Complexity is O(n)
Quick Sort:
The quick sort is considered to be a fast method to sort the elements. It was developed by
CAR Hoare. This method is based on divide-and-conquer technique i.e. the entire list is divided
into various partitions and sorting is applied again and again on these partitions. This method is
also called as partition exchange sorts.
The quick sort can be illustrated by the following example
12 6 18 4 9 8 2 15
Page 5 of 13
TutorialsD
uniya.com

The reduction step of the quick sort algorithm finds the final position of one of the
numbers. In this example, we use the first number, 12, which is called the pivot (rotate) element.
This is accomplished as follows-
Let ‘i’ be the position of the second element and ‘j’ be the position of the last element.
i.e. i =2 and j =8, in this example.
Assume that a [n+1] =, where ‘a’ is an array of size n.
12 6 18 4 9 8 2 15
[1] [2] [3] [4] [5] [6] [7] [8] [9]

i j
2 8
First scan the list from left to right (from i to j) can compare each and every element with
the pivot. This process continues until an element found which is greater than or equal to pivot
element. If such an element found, then that element position becomes the value of ‘i’.
Now scan the list from right to left (from j to i) and compare each and every element with
the pivot. This process continues until an element found which is less than or equal to pivot
element. If such an element finds then that element’s position become ‘j’ value.
Now compare ‘i’ and ‘j’. If i <j, then swap a[i] and a[j]. Otherwise swap pivot element
and a[j].
Continue the above process the entire list is sorted.
12 6 18 4 9 8 2 15
[1] [2] [3] [4] [5] [6] [7] [8] [9]

i j
2 8
12 6 18 4 9 8 2 15
12 6 2 4 9 8 18 15


3 7
7 6
Since i = 7 j=6, then swap pivot element and 6th element ( jth element), we get
8 6 2 4 9 12 18 15
Thus pivot reaches its original position. The elements on left to the right pivot are smaller
than pivot (12) and right to pivot are greater pivot (12).
8 6 2 4 9 12 18 15
Sublist 1 Sublist 2
Now take sub-list1 and sub-list2 and apply the above process recursively, at last we get
sorted list.
Ex 2: Let the given list is-
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10]
8 18 56 34 9 92 6 2 64
8 18 56 34 9 92 6 2 64
8 2 56 34 9 92 6 18 64
8 2 6 34 9 92 56 18 64




i j
2 98
2 8
3 7
4 3
8 18 56 34 9 92 6 2 64
Since i j, then swap jth element, and pivot element, we get
6 2
Sublist 1 Sublist 2
8 34 9 92 56 18 64
< > < >
Page 6 of 13
TutorialsD
uniya.com

Now take a sub-list that has more than one element and follow the same process as
above. At last, we get the sorted list that is, we get
2 6 8 9 18 34 56 64 92
The following algorithm shows the quick sort algorithm-
Algorithm Quicksort(i, j)
{
// sorts the array from a[i] through a[j]
If ( i <j) then //if there are more than one element
{
//divide P into two sub-programs
K: = partition (a, i, j+1);
//Here K denotes the position of the partitioning element
//solve the sub problems
Quicksort(i, K-1);
Quicksort(K=1, j);
// There is no need for combining solution
}
}
Algorithm Partition (a, left, right)
{
// The element from a[left] through a[right] are rearranged in such a manner that if initially
// pivot =a[left] then after completion a[j]= pivot, then return. Here j is the position where
// pivot partition the list into two partitions. Note that a[right]= .
pivot: a[left];
i:= left; j:=right;
repeat
{
repeat
i: =i+1;
until (a[i] ≥ pivot);
repeat
j: =j-1;
until (a[j] < pivot);
if( i<j) then
Swap (a, i, j);
}until (i ≥ j);
a[left]: = a[j];
a[j]: = pivot;
return j;
}
Algorithm Swap (a, i, j)
{
//Example a[i] with a[j]
temp:= a[i];
a[i]: = a[j];
a[j]:= temp;
}
Page 7 of 13
TutorialsD
uniya.com

Advantages of Quick-sort: Quick-sort is the fastest sorting method among all the sorting
methods. But it is somewhat complex and little difficult to implement than other sorting
methods.
Efficiency of Quick-sort: The efficiency of Quick-sort depends upon the selection of pivot
element.
Best Case: In best case, consider the following two assumptions-
1. The pivot, which we choose, will always be swapped into the exactly the middle of the
list. And also consider pivot will have an equal number of elements both to its left and
right.
2. The number of elements in the list is a power of 2 i.e. n= 2y
.
This can be rewritten as, y=log2n.
Thus, the total number of comparisons would be
O (n) + O (n) + O (n) +……..(y terms)
= O (n * y).
. .
.. .
.
Efficency in best case O( n log n) ( y=log2 n)
Page 8 of 13
TutorialsD
uniya.com

Worst Case: In worst case, assume that the pivot partition the list into two parts, so that one of
the partition has no elements while the other has all the other elements.
 Total number of comparisons will be-
 Thus, the efficiency of quick-sort in worst case is O (n2
).
Average Case: Let cA(n) be the average number of key comparisons made by quick-sort on a
list of elements of size n. assuming that the partitions split can happen in each position k(1kn)
With the same probability 1/n, we get the following recurrence relation.
Page 9 of 13
TutorialsD
uniya.com

The left child of each node represents a sub-problem size 1/4 as large, and the right child
represents a sub-problem size 3/4 as large.
There are log4/3 n levels, and so the total partitioning time is O(nlog4/3n). Now, there's a
mathematical fact that
logan = logbn / logba
for all positive numbers a, b, and n. Letting a=4/3 and b=2, we get that
log4/3 n=log n / log(4/3)
Quick Sort
Best Case O(n log n)
Average Case O(n log n)
Worst Case O(n2)
Space Complexity O(n)
Merge Sort:
Merge sort is based on divide-and-conquer technique. Merge sort method is a two phase
process-
1. Dividing
2. Merging
Dividing Phase: During the dividing phase, each time the given list of elements is divided into
two parts. This division process continues until the list is small enough to divide.
Merging Phase: Merging is the process of combining two sorted lists, so that, the resultant list is
also the sorted one. Suppose A is a sorted list with n element and B is a sorted list with n2
elements. The operation that combines the elements of A and B into a single sorted list C with
n=n1 + n2, elements is called merging.
Algorithm-(Divide algorithm)
Algorithm Divide (a, low, high)
{
// a is an array, low is the starting index and high is the end index of a
If( low < high) then
{
Mid: = (low + high) /2;
Divide( a, low, mid);
Divide( a, mid +1, high);
Merge(a, low, mid, high);
}
}
Page 10 of 13
TutorialsD
uniya.com

The merging algorithm is as follows-
Algorithm Merge( a, low, mid, high)
{
L:= low;
H:= high;
J:= mid +1;
K:= low;
While (low  mid AND j  high) do
{
If (a[low < a[j]) then
{
B[k] = a[low];
K:= k+1;
Low:= low+1;
}
Else
{
B[k]= a[j];
K: = k+1;
J: = j+1;
}
}
While (low  mid) do
{
B[k]=a[low];
K: = k+1;
Low: =low + 1;
}
While (j  high) do
{
B[k]=a[j];
K: = k+1;
j: =j + 1;
}
//copy elements of b to a
For i: = l to n do
{
A[i]: =b[i];
}
}
Page 11 of 13
TutorialsD
uniya.com

Ex: Let the list is: - 500, 345, 13, 256, 98, 1, 12, 3, 34, 45, 78, 92.
500 345 13 256 98 1 12 3 34 45 78 92
500 345 13 256 98 1 12 3 34 45 78 92
500 345 13 256 98 1
500 345 13 256 98 1
500 345 13 256 98 1
12 3 34 45 78 92
12 3 34
12 3 34
45 78 92
45 78 92
345 500 98 25613 1
13 345 500 1 98 256
500 345 13 256 98 1
3 12 45 7834 92
3 12 34 45 78 92
3 12 34 45 78 92
1 3 12 13 34 45 78 92 98 256 354 500 Sorted List
The merge sort algorithm works as follows-
Step 1: If the length of the list is 0 or 1, then it is already sorted, otherwise,
Step 2: Divide the unsorted list into two sub-lists of about half the size.
Step 3: Again sub-divide the sub-list into two parts. This process continues until each element in
the list becomes a single element.
Step 4: Apply merging to each sub-list and continue this process until we get one sorted list.
Efficiency of Merge List: Let ‘n’ be the size of the given list/ then the running time for merge
sort is given by the recurrence relation.
T(n) =
a if n=1, a is a constant
2T(n/2) + Cn if n>1, C is constant
{
Assume that ‘n’ is a power of 2 i.e. n=2k
.
This can be rewritten as k=log2 n.
Let T(n) = 2T (n/2) +Cn 1
We can solve this equation by using successive substitution.
Page 12 of 13
TutorialsD
uniya.com

1Replace n by n/2 in equation, ,we get
T (n/2) = 2T(n/4) + Cn
2
2
Thus, T(n) = 2 2T (n/4) + Cn + Cn
2
= 4T (n/4) + 2Cn
= 4T 2 T (n/8) + Cn + 2Cn
4
...
...
...
= 2 k T(1) + KCn
= a n + Cn log n
. .
.
T (n) = O( n log n)
. .
. k = log2 n
Worst case O(n log n)
Best case O(n log n)
Average case O(n log n)
Space Complexity О(2n)
Page 13 of 13
TutorialsD
uniya.com

UNIT-II1 GREEDY METHOD
1. GENERAL METHOD
Greedy method: It is most straight forward method. It is popular for obtaining the
optimized solutions.
Optimization Problem: An optimization problem is the problem of finding the best
solution (optimal solution) from all the feasible solutions (practicable of possible solutions).
In an optimization problem we are given a set of constraints and an optimization functions.
Solutions that satisfy the constraints are called feasible solutions. A feasible solution for
which the optimization function has the best possible value is called optimal solution.
Ex: Problem: Finding a minimum spanning tree from a weighted connected directed
graph G.
Constraints: Every time a minimum edge is added to the tree and adding of an edge
does not form a simple circuit.
Feasible solutions: The feasible solutions are the spanning trees of the given graph G.
Optimal solution: An optimal solution is a spanning tree with minimum cost i.e.
minimum spanning tree.
Q: Find the minimum spanning tree for the following graph.
The feasible solutions are the spanning tree of the graph G. Those are
From the above spanning tree the figure 4 gives the optimal solution, because it is the
spanning tree with the minimum cost i.e. it is a minimum spanning tree of the graph G.
The greedy technique suggests constructing a solution to an optimization problem
hrough a sequence of steps, each expanding a partially constructed solution obtained so far
until a complete solution to the problem is reached to each step, the choice made must be
feasible, locally optimal and irrecoverable.
Feasible: The choice which is made has to be satisfying the problems constraints.
Locally optimal: The choice has to be the best local choice among all feasible choices
available on that step.
Irrecoverable: The choice once made cannot be changed on sub-sequent steps of the
algorithm (Greedy method).
A B
C D
2
3 1
2
Graph G
A B
C D
2
3 1
A B
C D
3 1
2
A B
C D
2
3
2
A B
C D
2
1
2
1 .Total Weights=6 2 .Total Weights=6 3 .Total Weights=7 4 .Total Weights=5
Page 1 of 16
TutorialsD
uniya.com

Control Abstraction for Greedy Method:
Algorithm GreedyMethod (a, n)
{
// a is an array of n inputs
Solution: =Ø;
for i: =0 to n do
{
s: = select (a);
if (feasible (Solution, s)) then
{
Solution: = union (Solution, s);
}
else
reject (); // if solution is not feasible reject it.
}
return solution;
}
In greedy method there are three important activities.
1. A selection of solution from the given input domain is performed, i.e. s:= select(a).
2. The feasibility of the solution is performed, by using feasible ‘(solution, s)’ and then
all feasible solutions are obtained.
3. From the set of feasible solutions, the particular solution that minimizes or maximizes
the given objection function is obtained. Such a solution is called optimal solution.
Q: A child buys a candy 42 rupees and gives a 100 note to the cashier. Then the cashier
wishes to return change using the fewest number of coins. Assume that the cashier has Rs.1,
Rs. 5 and Rs. 10 coins.
This problem can be solved using the greedy method.
2. APPLICATION - JOB SEQUENCING WITH DEADLINES
This problem consists of n jobs each associated with a deadline and profit and our
objective is to earn maximum profit. We will earn profit only when job is completed on or
before deadline. We assume that each job will take unit time to complete.
Points to remember:
 In this problem we have n jobs j1, j2, … jn, each has an associated deadlines are d1,
d2, … dn and profits are p1, p2, ... pn.
 Profit will only be awarded or earned if the job is completed on or before the
deadline.
 We assume that each job takes unit time to complete.
 The objective is to earn maximum profit when only one job can be scheduled or
processed at any given time.
Page 2 of 16
TutorialsD
uniya.com

Example: Consider the following 5 jobs and their associated deadline and profit.
index 1 2 3 4 5
JOB j1 j2 j3 j4 j5
DEADLINE 2 1 3 2 1
PROFIT 60 100 20 40 20
Sort the jobs according to their profit in descending order.
Note! If two or more jobs are having the same profit then sorts them as per their entry
in the job list.
index 1 2 3 4 5
JOB j2 j1 j4 j3 j5
DEADLINE 1 2 2 3 1
PROFIT 100 60 40 20 20
Find the maximum deadline value
Looking at the jobs we can say the max deadline value is 3. So, dmax = 3
As dmax = 3 so we will have THREE slots to keep track of free time slots. Set the
time slot status to EMPTY
time slot 1 2 3
status EMPTY EMPTY EMPTY
Total number of jobs is 5. So we can write n = 5.
Note!
If we look at job j2, it has a deadline 1. This means we have to complete job j2 in time
slot 1 if we want to earn its profit.
Similarly, if we look at job j1 it has a deadline 2. This means we have to complete job
j1 on or before time slot 2 in order to earn its profit.
Similarly, if we look at job j3 it has a deadline 3. This means we have to complete job
j3 on or before time slot 3 in order to earn its profit.
Our objective is to select jobs that will give us higher profit.
time slot 1 2 3
Job J1 J2 J4
Profit 100 60 20
Total Profit is 180
Page 3 of 16
TutorialsD
uniya.com

Pseudo Code:
for i = 1 to n do
Set k = min(dmax, DEADLINE(i)) //where DEADLINE(i) denotes deadline of ith job
while k >= 1 do
if timeslot[k] is EMPTY then
timeslot[k] = job(i)
break
endif
Set k = k - 1
endwhile
endfor
Algorithm:
Time Complexity = O (n2
)
3. APPLICATION - KNAPSACK PROBLEM
In this problem the objective is to fill the knapsack with items to get maximum benefit
(value or profit) without crossing the weight capacity of the knapsack. And we are also
allowed to take an item in fractional part.
Points to remember:
In this problem we have a Knapsack that has a weight limit W
There are items i1, i2, ..., in each having weight w1, w2, … wn and some benefit
(value or profit) associated with it v1, v2, ..., vn
Our objective is to maximise the benefit such that the total weight inside the knapsack
is at most W. And we are also allowed to take an item in fractional part.
Page 4 of 16
TutorialsD
uniya.com

Example: Assume that we have a knapsack with max weight capacity, W = 16.
Our objective is to fill the knapsack with items such that the benefit (value or profit) is
maximum.
Consider the following items and their associated weight and value
ITEM WEIGHT VALUE
i1 6 6
i2 10 2
i3 3 1
i4 5 8
i5 1 3
i6 3 5
Steps
1. Calculate value per weight for each item (we can call this value density)
2. Sort the items as per the value density in descending order
3. Take as much item as possible not already taken in the knapsack
Compute density = (value/weight)
ITEM WEIGHT VALUE DENSITY
i1 6 6 1.000
i2 10 2 0.200
i3 3 1 0.333
i4 5 8 1.600
i5 1 3 3.000
i6 3 5 1.667
Sort the items as per density in descending order
ITEM WEIGHT VALUE DENSITY
i5 1 3 3.000
Page 5 of 16
TutorialsD
uniya.com

i6 3 5 1.667
i4 5 8 1.600
i1 6 6 1.000
i3 3 1 0.333
i2 10 2 0.200
Now we will pick items such that our benefit is maximum and total weight of the
selected items is at most W.
Our objective is to fill the knapsack with items to get maximum benefit without
crossing the weight limit W = 16.
How to fill Knapsack Table?
is WEIGHT(i) + TOTAL WEIGHT <= W
if its YES
then we take the whole item
How to find the Benefit?
If an item value is 10 and weight is 5
And if you are taking it completely
Then,
benefit = (weight taken) x (total value of the item / total weight of the item)
weight taken = 5 (as we are taking the complete (full) item, no fraction)
total value of the item = 10
total weight of the item = 5
So, benefit = 5 x (10/5) = 10
On the other hand if you are taking say, 1/2 of the item
Then,
weight taken = 5 x (1/2) = 5/2 (as we are taking 1/2 item)
So, benefit = (weight taken) x (total value of the item / total weight of the item)
= (5/2) x (10/5)
= 5
Values after calculation
ITEM WEIGHT VALUE TOTAL WEIGHT TOTAL BENEFIT
i5 1 3 1.000 3.000
i6 3 5 4.000 8.000
Page 6 of 16
TutorialsD
uniya.com

i4 5 8 9.000 16.000
i1 6 6 15.000 22.000
i3 1 0.333 16.000 22.333
So, total weight in the knapsack = 16 and total value inside it = 22.333336
Algorithm:
)
4. APPLICATION - MINIMUM SPANNING TREE
A spanning tree is a subset of Graph G, which has all the vertices covered with
minimum possible number of edges. Hence, a spanning tree does not have cycles and it
cannot be disconnected.
Note: Every connected and undirected Graph G has at least one spanning tree. A
disconnected graph does not have any spanning tree.
Page 7 of 16
TutorialsD
uniya.com

We found three spanning trees off one complete graph. A complete undirected graph
can have maximum nn-2
number of spanning trees, where n is the number of nodes. In the
above addressed example, 33−2
= 3 spanning trees are possible.
General Properties of Spanning Tree
 A connected graph G can have more than one spanning tree.
 All possible spanning trees of graph G, have the same number of edges and
vertices.
 The spanning tree does not have any cycle (loops).
 Removing one edge from the spanning tree will make the graph disconnected,
i.e. the spanning tree is minimally connected.
 Adding one edge to the spanning tree will create a circuit or loop, i.e. the
spanning tree is maximally acyclic.
Mathematical Properties of Spanning Tree
 Spanning tree has n-1 edges, where n is the number of nodes (vertices).
 From a complete graph, by removing maximum e - n + 1 edges, we can
construct a spanning tree.
 A complete graph can have maximum nn-2 number of spanning trees.
Thus, we can conclude that spanning trees are a subset of connected Graph G and
disconnected graphs do not have spanning tree.
Application of Spanning Tree
Spanning tree is basically used to find a minimum path to connect all nodes in a
graph. Common applications of spanning trees are
 Civil Network Planning
 Computer Network Routing Protocol
 Cluster Analysis
Let us understand this through a small example. Consider, city network as a huge
graph and now plans to deploy telephone lines in such a way that in minimum lines we can
connect to all city nodes. This is where the spanning tree comes into picture.
Minimum Spanning Tree (MST)
In a weighted graph, a minimum spanning tree is a spanning tree that has minimum
weight than all other spanning trees of the same graph. In real-world situations, this weight
can be measured as distance, congestion, traffic load or any arbitrary value denoted to the
edges.
Minimum Spanning-Tree Algorithm
We shall learn about two most important spanning tree algorithms(greedy
algorithms):
Page 8 of 16
TutorialsD
uniya.com

1. Kruskal's Algorithm
2. Prim's Algorithm
i. Kruskal's Algorithm
Kruskal's algorithm to find the minimum cost spanning tree uses the greedy
approach. This algorithm treats the graph as a forest and every node it has as an individual
tree. A tree connects to another only and only if, it has the least cost among all available
options and does not violate MST properties.
To understand Kruskal's algorithm let us consider the following example:
Step 1 - Remove all loops and Parallel Edges
Remove all loops and parallel edges from the given graph.
In case of parallel edges, keep the one which has the least cost associated and remove all
others.
Page 9 of 16
TutorialsD
uniya.com

Step 2 - Arrange all edges in their increasing order of weight
The next step is to create a set of edges and weight, and arrange them in an ascending order
of weightage (cost).
Step 3 - Add the edge which has the least weightage
Now we start adding edges to the graph beginning from the one which has the least weight.
Throughout, we shall keep checking that the spanning properties remain intact. In case, by
adding one edge, the spanning tree property does not hold then we shall consider not to
include the edge in the graph.
The least cost is 2 and edges involved are B,D and D,T. We add them. Adding them
does not violate spanning tree properties, so we continue to our next edge selection.
Next cost is 3, and associated edges are A,C and C,D. We add them again −
Next cost in the table is 4, and we observe that adding it will create a circuit in the graph. −
Page 10 of 16
TutorialsD
uniya.com

We ignore it. In the process we shall ignore/avoid all edges that create a circuit.
We observe that edges with cost 5 and 6 also create circuits. We ignore them and move on.
Now we are left with only one node to be added. Between the two least cost edges available
7 and 8, we shall add the edge with cost 7.
By adding edge S,A we have included all the nodes of the graph and we now have minimum
cost spanning tree.
Page 11 of 16
TutorialsD
uniya.com

Time Complexity = O (|E| log |E|)
ii. Prim’s Algorithm
Prim's algorithm to find minimum cost spanning tree (as Kruskal's algorithm) uses
the greedy approach. Prim's algorithm shares a similarity with the shortest path
first algorithms.
Prim's algorithm, in contrast with Kruskal's algorithm, treats the nodes as a single tree
and keeps on adding new nodes to the spanning tree from the given graph. To contrast with
Kruskal's algorithm and to understand Prim's algorithm better, we shall use the same
example.
Step 1 - Remove all loops and parallel edges
Page 12 of 16
TutorialsD
uniya.com

Remove all loops and parallel edges from the given graph. In case of parallel edges,
keep the one which has the least cost associated and remove all others.
Step 2 - Choose any arbitrary node as root node
In this case, we choose S node as the root node of Prim's spanning tree. This node is
arbitrarily chosen, so any node can be the root node. One may wonder why any video can be
a root node. So the answer is, in the spanning tree all the nodes of a graph are included and
because it is connected then there must be at least one edge, which will join it to the rest of
the tree.
Step 3 - Check outgoing edges and select the one with less cost
After choosing the root node S, we see that S,A and S,C are two edges with weight 7
and 8, respectively. We choose the edge S,A as it is lesser than the other.
Now, the tree S-7-A is treated as one node and we check for all edges going out from
it. We select the one which has the lowest cost and include it in the tree.
Page 13 of 16
TutorialsD
uniya.com

After this step, S-7-A-3-C tree is formed. Now we'll again treat it as a node and will
check all the edges again. However, we will choose only the least cost edge. In this case, C-
3-D is the new edge, which is less than other edges' cost 8, 6, 4, etc.
After adding node D to the spanning tree, we now have two edges going out of it
having the same cost, i.e. D-2-T and D-2-B. Thus, we can add either one. But the next step
will again yield edge 2 as the least cost. Hence, we are showing a spanning tree with both
edges included.
We may find that the output spanning tree of the same graph using two different
algorithms is same.
Page 14 of 16
TutorialsD
uniya.com

)
5. APPLICATION - SINGLE SOURCE SHORTEST PATH PROBLEM
For a given source node in the graph, the algorithm finds the shortest path between
that node and every other. It also used for finding the shortest paths from a single node to a
single destination node by stopping the algorithm once the shortest path to the destination node
has been determined.
Algorithm Steps:
 Set all vertices distances = infinity except for the source vertex, set the source
distance = 0.
 Push the source vertex in a min-priority queue in the form (distance , vertex), as the
comparison in the min-priority queue will be according to vertices distances.
 Pop the vertex with the minimum distance from the priority queue (at first the popped
vertex = source).
 Update the distances of the connected vertices to the popped vertex in case of "current
vertex distance + edge weight < next vertex distance", then push the vertex
with the new distance to the priority queue.
 If the popped vertex is visited before, just continue without using it.
 Apply the same algorithm again until the priority queue is empty.
Page 15 of 16
TutorialsD
uniya.com

Example:
Algorithm:
)
GREEDY APPROACH DIVIDE AND CONQUER
1.Many decisions and sequences areguaranteed
and all the overlapping subinstancesare consid
ered.
1.Divide the given problem into many subprobl
ems.Find the individual solutions andcombine t
hem to get the solution for themain problem
2. Follows Bottom-up technique 2. Follows top down technique
3.Split the input at every possible pointsrather
than at a particular point
3.Split the input only at specific points (midpoi
nt), each problem is independent.
4. Sub problems are dependent on the main
Problem
4. Sub problems are independent on the main
Problem
5. Time taken by this approach is not that
much efficient when compared with DAC.
5. Time taken by this approach efficient when
compared with GA.
6.Space requirement is less when compared
DAC approach.
6.Space requirement is very much high when
compared GA approach.
Page 16 of 16
TutorialsD
uniya.com

UNIT-IV DYNAMIC PROGRAMMING
Dynamic Programming:- Dynamic programming, like the divide-and-conquer method, solves
problems by combining the solutions to sub problems. Dynamic programming is applicable when the sub
problems are not independent, that is, when sub problems share sub sub-problems. A dynamic-programming
algorithm solves every sub sub-problem just once and then saves its answer in a table, thereby avoiding the
work of re-computing the answer every time the sub sub-problem is encountered. Dynamic programming is
typically applied to optimization problems. The development of a dynamic-programming algorithm can be
broken into a sequence of four steps.
1. Characterize the structure of an optimal solution.
2. Recursively define the value of an optimal solution.
3. Compute the value of an optimal solution in a bottom-up fashion.
4. Construct an optimal solution from computed information.
The dynamic programming technique was developed by Bellman based upon the principle known as
principle of optimality. Dynamic programming uses optimal substructure in a bottom-up fashion. That is, we
first find optimal solutions to sub problems and, having solved the sub problems, we find an optimal solution
to the problem.
Application-I: Matrix chain multiplication:- Matrix chain multiplication is an example of dynamic
programming. We are given a sequence (chain) A1, A2, ..., An of n matrices to be multiplied, and we wish to
compute the product A1 x A2 x A3 x….x An. We can evaluate the expression using the standard algorithm for
multiplying pairs of matrices as a subroutine once we have parenthesized it to resolve all ambiguities in how
the matrices are multiplied together. A product of matrices is fully parenthesized if it is either a single matrix
or the product of two fully parenthesized matrix products, surrounded by parentheses. Matrix multiplication
is associative, and so all parenthesize yield the same product. For example, if the chain of matrices is A1, A2,
A3, A4, the product A1 x A2 x A3 x A4 can be fully parenthesized in five distinct ways:
(A1 (A2 (A3 A4))) ,
(A1 ((A2 A3) A4)) ,
((A1 A2) (A3 A4)) ,
((A1 (A2 A3)) A4) ,
(((A1 A2) A3) A4).
The way we parenthesize a chain of matrices can have a dramatic impact on the cost of evaluating the
product. Consider first the cost of multiplying two matrices. We can multiply two matrices A and B only if
they are compatible: the number of columns of A must equal the number of rows of B. If A is a m × n matrix
and B is a p ×q matrix, the resulting matrix C is a m × q matrix. The standard algorithm is given below
Algorithm Matrix_Mul(A, B)
{
if (n ≠ P) then Error "incompatible dimensions"
else
for i ← 1 to m do
for j ← 1 to q do
{
C[i, j] ← 0
for k ← 1 to n do
C[i, j] ← C[i, j] + A[i, k] * B[k, j] }
return C
}
The time to compute C is the number of multiplications which is mnq or mpq.
Page 1 of 17
TutorialsD
uniya.com

Example 1:- To illustrate the different costs incurred by different parenthesize of a matrix product, consider
the problem to find the product of three matrices A1, A2, A3 i.e. A1 * A2 * A3 of three matrices. Suppose that
the dimensions of the matrices are 10 × 100, 100 × 5, and 5 × 50, respectively. If we multiply according to
the parenthesization.
((A1 A2) A3) = 10 * 100 * 5 + 10 * 5 * 50 = 7500
(A1 (A2 A3)) = 10 * 100 * 50 +100 * 5 * 50 = 75,000
Thus, computing the product according to the first parenthesization is 10 times faster.
Definition:- The matrix-chain multiplication problem can be stated as follows: given a chain A1, A2, ...,An
of n matrices, where for i = 1, 2, ..., n, matrix Ai has dimension Pi-1 ×Pi, fully parenthesize the product A1 A2
An in a way that minimizes the number of scalar multiplications.
Note:- In the matrix-chain multiplication problem, we are not actually multiplying matrices. Our goal is only
to determine an order for multiplying matrices that has the lowest cost.
Solving the matrix-chain multiplication problem by dynamic programming
Step 1: The structure of an optimal parenthesization. Our first step in the dynamic-programming paradigm is
to find the optimal substructure and then use it to construct an optimal solution to the problem from optimal
solutions to sub problems. For the matrix-chain multiplication problem, we can perform this step as follows.
For example any parenthesization of the product Ai Ai+1 Aj must split the product between Ak and Ak+1 for
some integer k in the range i ≤ k < j. That is, for some value of k, we first compute the matrices Ai,k and Ak+1,j
and then multiply them together to produce the final product Ai,j. The cost of this parenthesization is thus the
cost of computing the matrix Ai,k, plus the cost of computing Ak+1,j, plus the cost of multiplying them
together.
Step 2: A recursive solution. Next, we define the cost of an optimal solution recursively in terms of the
optimal solutions to sub problems. For the matrix-chain multiplication problem, We can define m[i, j]
recursively as follows. If i = j, the matrix Ai,j = Ai, so that no scalar multiplications are necessary to compute
the product. Thus, Mij = 0 for i = j
= min { Mi,k + Mk + 1, j + Pi-1 Pk Pj } for i < j
i<=k<j
Step 3: Computing the optimal costs. We perform the third step of the dynamic-programming paradigm and
compute the optimal cost by using a tabular, bottom-up approach.
Step 4: Constructing an optimal solution. In the first level we compare M12 and M23. When M12 < M23, we
parenthesize the A1A2 in the product A1A2A3 i.e (A1A2) A3 and parenthesize the A2A3 in the product A1A2A3
i.e., A1 (A2 A3) when M12 >M23. This process is repeated until the whole product is parenthesized. The top
entry in the table i.e M13 gives the optimum cost of matrix chain multiplication.
Example:- Find an optimal parenthesization of a matrix-chain product whose dimensions are given in the
table below.
Page 2 of 17
TutorialsD
uniya.com

Solution:- Given
P0=5 , P1=4, P2=6, P3=2, P4=7
The Bottom level of the table is to be initialized.
Mi,j=0 where i = j
To compute Mij when i < j,
Mij = min { Mi,k + Mk + 1, j + Pi-1 Pk Pj } for i < j i<=k<j
Thus M12 = min{ M11+M22+P0P1P2} = 0 + 0 + 5 * 4 * 6 = 120
M23 = min{ M22+M33+P1P2P3} = 0 + 0 + 4 * 6 * 2 = 48
M34 = min{ M33+M44+P2P3P4} = 0 + 0 + 6 * 2 * 7 = 84
M13 = min{ M11+M23+P0P1P3 , M12+M33+P0P2P3 }
= min{0 + 48 + 5 * 4 * 2 , 120+ 0+ 5 * 6 * 2} = min{ 88, 180} = 88
M24 = min{ M22+M34+P1P2P4 , M23+M44+ P1P3P4 }
= min{0 + 84 + 4 * 6 * 7 , 48+ 0+ 4 * 2 * 7} = min{ 252, 104} = 104
M14 = min{ M11+M24+P0P1P4 , M12+M34+P0P2P4 , M13+M44+P0P3P4 }
= min{0 + 104 + 5 * 4 * 7 , 120+ 84+ 5 * 6 * 7 , 88 + 0 + 5 * 2 * 7}
= min{ 244,414, 158} = 158
In the first level when we compare M12 , M23 and M34. As M23=48 is minimum among three we parenthesize
the QR in the product PQRT i.e P(QR)T. In the second level when we compare M13 and M24. As M13=88 is
minimum among the two we parenthesize the P and QR in the product PQRT i.e (P(QR))T. Finally we
parenthesize the whole product i.e ((P(QR))T). The top entry in the table i.e M14 gives the optimum cost of
((P(QR))T).
Verification:- The chain of matrices is P, Q, R, T, the product P x Q x R x T can be fully parenthesized in
five distinct ways:
1. (P(Q(RT))) 2. (P((QR)T)) 3. ((PQ)(RT)) 4. ((P(QR))T) 5. (((PQ) R)T)
Cost of (P(Q(RT))) = 5*4*7 + 4*6*7 + 6*2*7 = 392
Cost of (P((QR)T)) = 5*4*7 + 4*6*2 + 4*2*7 = 244
Cost of ((PQ)(RT)) = 5*4*6 + 6*2*7 + 5*6*7 = 414
Cost of ((P(QR))T) = 5*4*2 + 4*6*2 + 5*2*7 = 158
Cost of (((PQ) R)T) = 5*4*6 + 5*6*2 + 5*2*7 = 250
Page 3 of 17
TutorialsD
uniya.com

Download FREE Computer Science Notes, Programs,
Projects, Books for any university student of BCA,
MCA, B.Sc, M.Sc, B.Tech CSE, M.Tech at
https://www.tutorialsduniya.com
Download Complete
Notes PDF for FREE
at
TutorialsDuniya.com
TUTORIALSDUNIYA.COM
Computer Science Notes

Algorithms notes 2 tutorials duniya

Recommended

More Related Content

What's hot (20)

Similar to Algorithms notes 2 tutorials duniya (20)

More from TutorialsDuniya.com (20)

Recently uploaded (20)

Algorithms notes 2 tutorials duniya