Definition 2.1.

1
An alphabet is a finite set of symbols.
Definition 2.1.2
A string over an alphabet is a finite sequence of symbols from the alphabet.
We usually use u, v, w, x, y, z and Greek letters to denote strings.
Definition 2.1.3
Alphabets and Languages The empty string, denoted by e, is the string containing no symbols at all.
Note:
Set of all strings, including empty string, over an alphabet  is denoted by *.
Example:
 = {a, b}
 * = {e, a, b, aa, ab, ba, bb, aaa, aab, aba, abb, baa, bab, … }
Definition 2.1.4
The length of a string w, denoted by w, is defined as its length as a sequence.
1
2

Definition 2.1.5 Definition 2.1.8

The concatenation, w, of two strings x and y over a same alphabet is defined as The reversal of a string w  *, denoted by wR, is defined inductively as
the string x followed by string y, and is denoted by w = x o y or w = xy. 1. If w= 0, then wR = w = e
Note: 2. If w > 0, then w = ua for some a  , u  *, and, wR = auR
• x o y = x + y
Note
• w o e = e o w = w for any string w
Reversal of a string is the string spell backwardly.
• Concatenation is associative. i.e. x o (y o z) = (x o y) o z for any strings x, y
and z 2.2 Languages
Definition 2.1.6 Definition 2.2.1
A string v is a substring of a string w if and only if there exists strings x and y Any set of strings over an alphabet  is called a language.
such that w = xvy.
Note
Definition 2.1.7
• *, , and  are languages over 
For each string w  *, w i is defined as follows
• * is infinite
w0=e
w i+1 = w i o w for each i  0 Operations on languages
Definition 2.2.2
The complement of a language L over , denoted by L , is defined as * - L.
3 4
Definition 2.2.3 Note
If L1 and L2 are languages over , their concatenation L, denoted by L • * = {e}
= L1 o L2 (or L1L2), is defined as
• For any languages L1, L2, L1  L2,  L1*  L2*
L = { w  * w = x o y for some x  L1, y  L2}
Definition 2.2.5
Example:
Let  = {0, 1}, L+ = L L* = {w  *w = w1 o … o wk; k  1, and some w1, …, wk  L}
L1 = { w  *w has an even no. of 0s} 2.3 Finite representation of languages
L2 = { w  *w starts with 0 and rest are 1s}  Finite language  Can list all the strings in the language
 L1 o L2 = {w  *w has an odd no. of 0s}  Infinite language (most languages)  Need to represent them finitely
Desired properties in representation
Definition 2.2.4
The Kleene star of a language L, denoted by L*, is the set of all strings  representation should be by a string, a sequence of symbols over an
obtained by concatenating zero or more strings from L. i.e. alphabet
L* = {w  *w = w1 o … o wk ; k  0, and some w1, …, wk  L}  different languages should have different representations
Example 2.2.3
Let L = {00, 01, 10, 11}
 1000111110 = 10 o 00 o 11 o 11 o 10  L* 5 6

Note Definition 2.4.1

• Set * of strings over an alphabet  is countably infinite
The regular expressions over an alphabet  are all strings over alphabet
no. of possible representations is countable   {(, ), , , *} satisfying the following properties
• 2*, set of all possible languages over an alphabet , is uncountable
•  is a regular expression
 We cannot represent all languages finitely • Every member of  is a regular expression
2.4 Regular languages • If  and  are regular expressions, then so are (), (  ), and *

Consider the language • Every regular expression represents a language

L = { w  {0,1}* w has 2 or 3 occurrences of 1, the 1st and 2nd of which are • Relation between regular expressions and the language they represent is
not consecutive} established by a function L
L can be described by using only singleton sets and language operators; L() - language represented by the regular expression 
concatenation, Kleene star, and union (set union):
L = {0}* o {1} o {0}* o {0} o {1} o {0}* o (({1} o {0}*)  *)
= 0*10*010* (10*  *)

Regular expression 7
8
Definition 2.4.2 Example:
• L() =  Let  = c*(a  (bc*))*
• L(a) = {a} for any a    L() is the set of all strings over {a, b, c} that do not contain the
substring ac
• If  and  are regular expressions, then
L() = L() L() Definition 2.4.3
L(  ) = L()  L() Class of regular languages over an alphabet  is defined as the set of all
languages L() for some regular expression  over .
L(*) = L()*
Example: Regular languages can be defined in terms of closures.
Let  {a, b}. The language represented by the regular expression (a  b)*a Class of regular languages over  is the closure of the set of languages
is
{ {}     }  {} under union, concatenation, and Kleene star.
L((a  b)*a) = L((a  b)*) L(a)
i.e., the class of regular languages is the set of languages satisfying the
= L((a  b)*) {a}
following properties:
= L(a  b)* {a}
= (L(a)  L(b))* {a} • Empty language () is regular
= ({a}  {b})* {a} = {a, b}* {a} • {} ;    is a regular language
= {w  {a, b} w ends with an a} • If L1 and L2 are regular languages, then L1*, L1  L2, and L1 o L2
9 are also regular 10

Note Alternative method (language generators)

• Not possible to describe all languages by regular expressions
Ex: {0n1n  n  0} is not regular Example:
• Regular expressions are an inadequate method of language specification Let  = (e  b  aa) (a  ab  aab)*
Language recognition device Members of L() can be generated as:
An algorithm that is designed to check whether a certain string is a member • Write down either nothing, or b, or aa
of the language. • Write down a or ab, or aab, and do this any number of times,
Example: including zero
L = {w  {0, 1}* w does not have 111 as a string} Note
A device for recognizing L Such language generators are not algorithms.
• Read strings, one symbol at a time, from left to right
• Keep a count (initially zero)
• Add 1 to count when a 1 is encountered in input. Set count to zero
when a 0 is encountered
• Stop with answer “No” if count reaches three, or stop with answer
“Yes” if the whole string is read without count reaching the value of
three 11 12