Lecture # 5

CS-312
Theory of Automata and Formal
Languages
 Agenda
Regular Expression
 Formal Definition of Regular Expression
Regular Languages
 Languages associated with regular expressions
 All finite languages are regular
Introduction to Finite Automata:
 States
 Transition
 Acceptance or Rejection of a string
 Representation of an FA by Transition Table
2
 Regular Expression

3
 Regular Expression
 Given  = {a,b}
 a* = {Λ, a,aa,aaa,aaa,aaaa,aaaaa, …}
 ab* = {a, ab,abb,abbb,abbbb, …}
 a+b = {a,b}
 (ab)* = {Λ, ab, abab, ababab, …}
 (a+b)* = {Λ, any string of a’s and b’s}

4
 Formal Definition of Regular
Expressions
The set of regular expression is defined by following rules
1. Every letter of  and Λ is a regular expression
2. If r1 and r2 are regular expressions, then so are
 r 1r 2
 r1+r2
 r1* (or r2* )

3.Nothing else is a regular expression

5
 Example of RE
 Now consider another language L, of even length, defined
over Σ = {a, b}, then it’s regular expression may be
 (a+b) generate 1 letter which is of length 1
 (a+b) (a+b) generate 2 letter which is even length
 ( (a+b) (a+b) ) * generate language of even length
Generate even length string where length =4

((a+b) (a+b)) * ((a+b) (a+b)) *

((a+b) (a+b)) ((a+b) (a+b)) ((a+b) (a+b)) ((a+b) (a+b))
( ( a ) ( b )) (( b ) ( b )) ( ( a ) ( a )) (( a ) ( a ))
abbb aaaa

6
 Example of RE
 Now consider another language L, of odd length,
defined over Σ = {a, b}, then it’s regular expression
may be
(a+b)((a+b)(a+b))* or
((a+b)(a+b))*(a+b)

7
 Remark
 It may be noted that a language may be expressed
by more than one regular expressions,
while given a regular expression there exist a
unique language generated by that regular
expression.

8
 Example of RE
 Consider the language, defined over
Σ={a, b}, of words starting with double a and
ending in double b then its regular expression may
be aa(a+b)*bb
 Consider the language, defined over
Σ={a, b} of words starting with a and ending in
b OR starting with b and ending in a, then its
regular expression may be a(a+b)*b+b(a+b)*a

9
 Example of RE

 Example:
 Consider the language, defined over
Σ={a , b} of words having at least one a, may be
expressed by a regular expression
(a+b)*a(a+b)*.
 Consider the language, defined over
Σ = {a, b} of words having at least one a and one
b, may be expressed by a regular expression
(a+b)*a(a+b)*b(a+b)*+ (a+b)*b(a+b)*a(a+b)*.

10
 Example of RE
 Consider the language, defined over
Σ={a, b} of words beginning with a, then its
regular expression may be a(a+b)*
 Consider the language, defined over
Σ={a, b} of words beginning and ending in
same letter, then its regular expression may be
 a(a+b)*a
 b(a+b)*b
 (a+b)
 (a+b)+a(a+b)*a+b(a+b)*b

11
 Example of RE
 Consider the language, defined over
Σ={a, b} of words ending in b, then its regular
expression may be (a+b)*b.
 Consider the language, defined over
Σ={a, b} of words not ending in a, then its regular
expression may be (a+b)*b + Λ. It is to be noted that this
language may also be expressed by ((a+b)*b)*.

12
 An important example
The Language EVEN-EVEN :
Language of strings, defined over Σ={a, b} having even
number of a’s and even number of b’s. i.e.
EVEN-EVEN = {Λ, aa, bb, aaaa,aabb,abab, abba, baab, baba,
bbaa, bbbb,…} ,
its regular expression can be written as
(aa + bb + (ab+ba)(aa+bb)*(ab+ba))*
To Understand:
(ab+ba)(aa+bb)*(ab+ba) = (ab)(aa)(ba) =abaaba
(ab+ba)(aa+bb)*(ab+ba) = (ab)(bb)(ba) =abbbba
13
 Note
 It is important to be clear about the difference of
the following regular expressions
r1=a*+b*
r2=(a+b)*
Here r1 does not generate any string of
concatenation of a and b, while r2 generates such
strings.

14
 Equivalent Regular Expressions
 Definition:
Two regular expressions are said to be equivalent if they
generate the same language.
Example:
Consider the following regular expressions
r1= (a + b)* (aa + bb)
r2= (a + b)*aa + ( a + b)*bb then
both regular expressions define the language of strings
ending in aa or bb.

15
 Regular Languages

16
 Regular Languages Definition
 The language generated by any regular expression is
called a regular language.

 It is to be noted that if r1, r2 are regular expressions ,

corresponding to the languages L1 and L2 then the
languages generated by r1+ r2, r1r2( or r2r1) and
 r1*( or r2*) are also regular languages.

17
 Note
 It is to be noted that if L1 and L2 are expressed by r1and r2,
respectively then the language expressed by
1) r1+ r2, is the language L1 + L2 or L1 U L2
2) r1r2, , is the language L1L2, of strings obtained by prefixing
every string of L1 with every string of L2
3) r1*, is the language L1*, of strings obtained by concatenating the
strings of L, including the null string.

18
 Example
 If r1=(aa+bb) and r2=(a+b) then the language of strings generated
by r1+r2, is also a regular language, expressed by (aa+bb)+(a+b)
 If r1=(aa+bb) and r2=(a+b) then the language of strings
generated by r1r2, is also a regular language, expressed by
(aa+bb)(a+b)
 If r=(aa+bb) then the language of strings generated by r*, is also a
regular language, expressed by (aa+bb)*

19
 All finite languages are regular
Example:
Consider the language L, defined over Σ={a,b}, of strings of
length 2, starting with a, then
L={aa, ab}, may be expressed by the regular expression
aa+ab. Hence L, by definition, is a regular language.

20
 Note
 It may be noted that if a language contains even thousand
words, its RE may be expressed, placing ‘ + ’ between all the
words.
 Here the special structure of RE is not important.
 Consider the language
L={aaa, aab, aba, abb, baa, bab, bba, bbb},
that may be expressed by a
RE=aaa+aab+aba+abb+baa+bab+bba+bbb,
which is equivalent to
(a+b)(a+b)(a+b).

21
 What are Regular Language of
following Regular Expressions?
 Whether following are RE if so what languages do they
generate
 a (b + a)*
 bb(a+b)
 (a+b)(a+b)(a+b)
 (a+b)*ba
 (a+b)*a(a+b)*
 (a+b)*aa(a+b)*

22
 What are Regular Expressions of
following Regular expression?
 Write RE for the following languages over the  ={a,b}.
 All words ending with b
 All words that start with a
 All words that start with a double letter
 All words that contain at least one double letter
 All words that start and end with a double letter
 All words of length >=3
 All words that contain exactly one a or exactly one b
 All words that don’t end at b

23
 Finite Automata

24
 Introduction to Defining Languages

Languages can be defined in different ways , such as

 Descriptive definition
 Recursive definition
 Using Regular Expressions(RE)
 Using Finite Automaton(FA)

25
 Finite Automata
 Language Recognizers
 Machines embedded with grammatical rules that recognize a
language
 REs define a language and FAs accept (or reject ) them

 Visual notations (abstract machines)

26
 Finite Automata
 Sort of graphs consisting of nodes called states and edges

called transitions

 States serve as memory locations that keep a track of last

character read

 Transitions define where to go on reading a particular

character

27
 Finite Automata
 A finite automaton is a collection of followings:
 A finite set of states
 Exactly one initial state (start state)
 One or more (may be none) final states that mark the
acceptance of a word
 Intermediate states that are neither start not final states
 An alphabet  of possible input letters

28
 How Does a Finite Automaton
work?
 The start state marks the beginning of reading every
input
 Reading a character triggers a transition from that
state which may transfer control to some other state and
the reading mechanism advances to next character and
the process continues
 When the input terminates, if the control is left
with a final or accepting state, the input string is
accepted otherwise it is rejected and the FA resets
control to the initial state for next input
29
 How Does a Finite Automaton
work?
 The state to go to next on reading a letter of the input

string is determined automatically and deterministically

by the transitions and is fixed (for that particular state

and input character)

 The transitions are fired automatically

 A single character is consumed for each transition

30
 Finite Automata Visual
Representation
 Visual representations
 States represented by circles labeled to identify each distinctly

1/x
 Initial (- sign) and Final states (+ sign )

- +
 Transitions
 Directed edges labeled with the characters of 

a
- +

31
 Example of Finite Automata
Consider the following FA:
The input alphabet has only the two letters a and b. (We usually use
this alphabet throughout the chapter.)

There are only three states, x, y and z, where x is the start state and
z is the final state.
Input Alphabets

The transition list for this FA is as follows: State a b

– Rule 1: From state x and input a, go to state y.
– Rule 2: From state x and input b, go to state z. x y z
– Rule 3: From state y and input a, go to state x.
y x z
– Rule 4: From state y and input b, go to state z.
32 – Rule 5: From state z and any input, stay at state z. z z z
 Finite Automata
a

x y
Input a
Alphabets

State a b
x y z
y x z b b
z z z a b
33
 Example of Finite Automata
Let us examine what happens when the input string aaa is
presented to this FA.

First input a: state x → y by Rule 1.

Second input a: state y → x by Rule 3.
Third input a: state x → y by Rule 1.

We did not finish up in the final state z, and therefore

have an unsuccessful termination.

34
 Example of Finite Automata
The set of all strings that lead to a final state is called the
language defined by the finite automaton.
Thus, the string aaa is not in the language defined by
this FA.
We may also say that the string aaa is not accepted by this
FA, or the string aaa is rejected by this FA.
The set of all strings accepted is also called the language
associated with the FA.
We also say, “This FA accepts the language L”, or “L is the
language accepted by this FA”, or “L is the language of
the FA”, by which we mean that all the words in L are
accepted, and all the inputs accepted are words in L

35
 Example of Finite Automata
It is not difficult to find the language accepted by this FA.
If an input string is made up of only letter a’s then the action
of the FA will be to jump back and forth between state x and
state y.
To get to state z, it is necessary for the string to have the letter
b in it. As soon as a b is encountered, the FA jumps to state z.
Once in state z, it is impossible to leave. When the input string
runs out, the FA will be in the final state z.
This FA will accept all strings that have the letter b in them.
Hence, the language accepted by this FA is defined by the
regular expression
36 (a + b)*b(a + b)*
 Abstract definition of FA
1. FA = (Q, ,q0,F,δ)
2. A finite set of states Q = {q0, q1, q2 q3 … qn where n is
finite }
3. An alphabet ∑ = { x1, x2, x3, …}.
4. q0 is the start states
5. F  Q is the set of final states F may be 
6. A transition function δ associating each pair of state and
letter with a state:
δ(q,xj) = xk

37
 Transition Diagrams
Pictorial representation of an FA gives us more of a
feeling for the motion.
We represent each state by a small circle.
We draw arrows showing to which other states the
different input letters will lead us. We label these
arrows with the corresponding input letters.
If a certain letter makes a state go back to itself, we
indicate this by a loop.
We indicate the start state by a minus sign, or by
labeling it with the word start.
We indicate the final states by plus signs, or by labeling
them with the word final.
Sometimes, a start state is indicated by an arrow, and a
38
final state is indicated by drawing another circle around its
 FA with Minus and Plus Sign

39
 FA with start and final label

40
 FA with arrow and double circle

41
 Transition Diagram
When we depict an FA as circles and arrows, we say that
we have drawn a directed graph.
We borrow from Graph Theory the name directed
edge, or simply edge, for the arrow between states.
Every state has as many outgoing edges as there
are letters in the alphabet.
It is possible for a state to have no incoming
edges or to have many.

42
 Finite Automata Example
By convention, we say that the null string starts in the
start state and ends also in the start state for all FAs.
Consider this FA:

The language accepted by this FA is the set of all strings

except Ʌ.
The regular expression of this language is
(a + b)(a + b)* = (a + b)+
43
 Finite Automata Example
 One of many FAs that accept all words is

 Here, the ± means that the same state is both a start and a
final state.
 The language for this machine is
(a + b)*

44
 Finite Automata Example
 The second type include FAs of which the final states can not be
reached from the start state.

 This may be either because the diagram is in two separate

components. In this case, we say that the graph is disconnected,
as in the example below:

45
 Finite Automata Example
 Or it is because the final state has no incoming edges, as
shown below:

46
 FA and their Languages
We will study FA from two different angles:

1. Given a language, can we build a machine for it?

2. Given a machine, can we deduce its language?

47
 Example
Let us build a machine that accepts the language of all words over
the alphabet Σ = {a, b} with an even number of letters.
A mathematician could approach this problem by counting the total
number of letters from left to right. A computer scientist would
solve the problem differently since it is not necessary to do all the
counting:
Use a Boolean flag, named E, initialized with the value TRUE. Every
time we read a letter, we reverse the value of E until we have
exhausted the input string. We then check the value of E. If it is
TRUE, then the input string is in the language; if FALSE, it is not.
The FA for this language should require only 2 states:
– State 1: E is TRUE. This is the start and also final state.
– State 2: E is FALSE.

48
 Example Contd.
 So the FA is pictured as follows:

49
 Example
Let us build a FA of a language of all strings that begin with the
letter a.
a(a + b)*

Starting at state x, if we read a letter b, we go to a dead-end state

y. A dead-end state is one that no string can leave once it has
entered.

If the first letter we read is an a, then we go to the dead-end state z,

which is also a final state.

50
 Example
 The machine looks like this:

51
 Example
 Let’s build a machine that accepts all words containing a
triple letter, either aaa or bbb, and only those words.

 From the start state, the FA must have a path of three edges,
with no loop, to accept the word aaa. So, we begin our FA
with the following:

52
 Example Contd.
 For similar reason, there must be a path for bbb, that has no loop,
and uses entirely differently states. If the b-path shares any states
with the a-path, we could mix a’s and b’s to get to the final state.
However, the final state can be shared.

 So, our FA should now look like:

53
 Example Contd.
 If we are moving along the a-path and we read a b before the
third a, we need to jump to the b-path in progress and
vice versa.The final FA then looks like this:

54
 Example
 Consider the FA below. We would like to examine what language
this machine accepts.

55
 Example
There are only two ways to get to the final state 4 in this FA: One is from state 2
and the other is from state 3.

The only way to get to state 2 is by reading an a while in either state 1 or state 3.
If we read another a we will go to the final state 4.

Similarly, to get to state 3, we need to read the input letter b while in either state
1 or state 2. Once in state 3, if we read another b, we will go to the final state 4.

Thus, the words accepted by this machine are exactly those strings that have a
double letter aa or bb in them. This language is defined by the regular expression

(a + b)*(aa + bb)(a + b)*

56
 Example
 Consider the FA below. What is the language accepted by this
machine?

57
 Example
Starting at state 1, if we read a word beginning with an a, we will
go straight to the final state 3. We will stay in state 3 as long as we
continue to read only a’s. Hence, all words of the form aa are
accepted by this FA.

What if we began with some a’s that take us to state 3 and then we
read a b? This will bring us to state 2. To get back to the final state 3,
we must proceed to state 4 and then state 3. This trip requires two
more b’s.

Notice that in states 2, 3, and 4, all a’s that are read are ignored; and
only b’s cause a change of state.

58
 Summarizing what we know: If an input string starts with an a
followed by some b’s, then it must have 3 b’s to return to the final
state 3, or 6 b’s to make the trip twice, or 9 b’s, or 12 b’s and so on.

In other words, an input string starting with an a and having a total
number of b’s divisible by 3 will be accepted. If an input string
starts with an a but has a total number of b’s not divisible by 3, then
it is rejected because its path will end at either state 2 or 4.

59
 Example Contd.
What happens to an input string that begins with a b?

Such an input string will lead us to state 2. It then needs two more b’s to
get to the final state 3. These b’s can be separated by any number of a’s.
Once in state 3, it needs no more b’s, or 3 more b’s, or 6 more b’s and so
on.

All in all, an input string, whether starting with an a or a b, must have a

total number of b’s divisible by 3 to be accepted.

The language accepted by this machine therefore can be defined by the

regular expression

(a + ba*ba*b)+ = (a + ba*ba*b)(a + ba*ba*b)*

60
 Example EVEN-EVEN revisited
 Consider the FA below.

61
 Example EVEN-EVEN revisited Contd.
There are 4 edges labeled a. All the a-edges go either from one of the
upper two states (states 1 and 2) to one of the lower two states (states 3
and 4), or else form one of the lower two states to one of the upper two
states.

Thus, if we are north and we read an a, we go south. If we are south and

we read an a, we go north.

If a string gets accepted by this FA, we can say that the string must have
had an even number of a’s in it. Every a that took us south was balanced by
some a that took us back north.

So, every word in the language of this FA has an even number of a’s in it.
Also, we can say that every input string with an even number of a will
finish its path in the north (ie., state 1 or state 2).

62
 Example EVEN-EVEN revisited Contd.
Therefore, all the words in the language accepted by this FA must
have an even number of a’s and an even number of b’s. So, they are
in the language EVEN-EVEN.

Notice that all input strings that end in state 2 have an even number
of a’s but an odd number of b’s. All strings that end in state 3 have an
even number of b’s but an odd number of a’s. All strings that end in
state 4 have an odd number of a’s and an odd number of b’s. Thus,
every word in the language EVEN - EVEN must end in state 1 and
therefore be accepted.

Hence, the language accepted by this FA is EVEN-EVEN.

63