Normal forms for ContextFree Grammars

Context-Free Grammar

In linguistics and computer science, a context-free grammar (CFG) is a formal grammar in which every production rule is of the form Vw where V is a non-terminal symbol and w is a string consisting of terminals and/or non-terminals.

The term "context-free" expresses the fact that the non-terminal V can always be replaced by w, regardless of the context in which it occurs. A formal language is context-free if there is a context-free grammar that generates it.

Context-Free Grammar

Context-free grammars are powerful enough to describe the syntax of most programming languages; in fact, the syntax of most programming languages is specified using context-free grammars. On the other hand, context-free grammars are simple enough to allow the construction of efficient parsing algorithms which, for a given string, determine whether and how it can be generated from the grammar.

Context-Free Grammar

Not all formal languages are context-free. A well-known counter example is n bn cn : n >= 0 } {a the set of strings containing some number of a's, followed by the same number of b's and the same number of c's.

Context-Free Grammar

Just as any formal grammar, a context-free grammar G can be defined as a 4-tuple: G = (Vt ,Vn ,P,S) where Vt is a finite set of terminals Vn is a finite set of non-terminals P is a finite set of production rules S is an element of Vn, the distinguished starting non-terminal.

elements of P are of the form

Vn ( Vt U Vn) *

A language L is said to be a Context-Free-Language (CFL) if its grammar is Context-Free. More precisely, it is a language whose words, sentences and phrases are made of symbols and words from a Context-Free-Grammar. Usually, CFL is of the form L=L(G).

Example 1

A simple context-free grammar is given as:

SaSb|

where | is used to separate multiple options for the same non-terminal, and stands for the empty string. This grammar generates the language { an bn : n >= 0 } , which is not regular.

Regular languages

A regular language is a formal language (i.e., a possibly infinite set of finite sequences of symbols from a finite alphabet) that satisfies the following equivalent properties: it can be accepted by a deterministic finite state machine it can be accepted by a nondeterministic finite state machine it can be accepted by an alternating finite automaton it can be described by a regular expression it can be generated by a regular grammar it can be generated by a prefix grammar

Regular languages
The collection of regular languages over an alphabet is defined recursively as follows: the empty language is a regular language. the empty string language { } is a regular language. For each a , the singleton language { a } is a regular language. If A and B are regular languages, then A B (union), A B (concatenation), and A* (Kleene star) are regular languages. No other languages over are regular.

Finite languages
Finite languages are: A specific subset within the class of regular languages is the finite languages - those containing only a finite number of words. These are obviously regular as one can create a regular expression that is the union of every word in the language, and thus are regular.

Example 2

A context-free grammar for the language consisting of all strings over {a,b} which contain a different number of a's to b's is

SU|V U TaU | TaT V TbV | TbT T aTbT | bTaT |

Here, T can generate all strings with the same number of a's as b's, U generates all strings with more a's than b's and V generates all strings with fewer a's than b's.

Example 3

Another example of a context-free language is This is not a regular language, but it is context free as it can be generated by the following context-free grammar:

S b S bb | A AaA|

Normal forms

Every context-free grammar that does not generate the empty string can be transformed into an equivalent one in Chomsky normal form or Greibach normal form. "Equivalent" here means that the two grammars generate the same language. Because of the especially simple form of production rules in Chomsky Normal Form grammars, this normal form has both theoretical and practical implications. For instance, given a context-free grammar, one can use the Chomsky Normal Form to construct a polynomial-time algorithm which decides whether a given string is in the language represented by that grammar or not (the CYK algorithm).

Properties of context-free languages

An alternative and equivalent definition of contextfree languages employs non-deterministic pushdown automata: a language is context-free if and only if it can be accepted by such an automaton. A language can also be modeled as a set of all sequences of terminals which are accepted by the grammar. This model is helpful in understanding set operations on languages. The union and concatenation of two context-free languages is context-free, but the intersection need not be. The reverse of a context-free language is contextfree, but the complement need not be.

Properties of context-free languages

Every regular language is context-free because it can be described by a regular grammar. The intersection of a context-free language and a regular language is always context-free. There exist context-sensitive languages which are not context-free. To prove that a given language is not context-free, one may employ the pumping lemma for contextfree languages. The problem of determining if a context-sensitive grammar describes a context-free language is undecidable.

Normal forms for Context-Free Grammars

The goal is to show that every CFL (without ) is generated by a CFG in which all productions are of the form A BC or A a, where A, B, C are variables, and a is a terminal.

Normal forms for Context-Free Grammars

1.

2.

3.

A number of simplifications is inevitable: The elimination of useless symbols, variables or terminals that do not appear in any derivation of a terminal string from the start symbol. The elimination of -productions, those of the form A for some variable A. The elimination of unit productions, those of the form A B for variables A and B.

Eliminating useless symbols

A symbol X is useful for Grammar G = {V, T, P, S}, if there is some derivation of the form S >* a X b >* w , where w T* X V or X T The sentential form of a X b might be the first or last derivation If X is not useful, then X is useless

Eliminating useless symbols

Characteristics of useful symbols (for instance X): X is generating if X >* w for some terminal string w. Every terminal is generating since w can be that terminal itself, which is derived by 0 steps. X is reachable if there is a derivation S >* a X b for some a and b. A symbol which is useful is surely to be both generating and reachable.

1.

2.

Eliminating useless symbols

Eliminating the symbols which are not generating first followed by eliminating the symbols which are not reachable from the remaining grammar, this will generate a grammar consisting of only useful symbols.

Eliminating useless symbols

Example 7.1 If we have the following grammar:

Eliminating useless symbols

Example 7.1 Notice that a and b generate themselves terminals, S generates a, and A generates b. B is not generating. After eliminating B:

Eliminating useless symbols

Example 7.1 Notice that only S and a are reachable after eliminating the non-generating B. A is not reachable; so it should be eliminated. The result :

This production itself is a grammar that has the same result, which is {a}, as the original grammar.

Computing the generating and reachable symbols

Basis: Every Symbol of T is obviously generating; it generates itself. Induction: If we have a production A a, and every symbol of a is already known to be generating, then A is generating; because it generates all and only generating symbols, even if a = ; since all variables that have as a production body are generating. Theorem: The previous algorithm finds all and only the Generating symbols of G

Computing the generating and reachable symbols

Basis : For a grammar G = {V, T, P, S} S is surely reachable. Induction: If we discovered that some variable A is reachable, then for all productions with A in the head (first part of the expression), all the symbols of the bodies (second part of the expression) of those productions are also reachable. Theorem: The above algorithm finds all and only the Reachable symbols of G

Eliminating useless symbols

So far, the first step, which is the elimination of useless symbols is concluded. Now, for the second part, which is the elimination of -productions.

Eliminating -productions

The strategy is to have the following: if L is CFG, then L {} is also CFG This is done through discovering the nullable variables. A variable for instance A, is nullable if: A >* . Whenever A appears in a production body, A might or might not derive

Eliminating -productions

Basis: If A is a production of G, then A is nullable Induction: If there is a production B C1 C2 Ck such that each C is a variable and each C is nullable, then B is nullable

Eliminating -productions

Theorem: For any grammar G, the only nullable symbols are the variables that derive in previous algorithm Proof: for one step : A must be a production, then this implies that A is discovered as nullable (as in basis). for N > 1 steps: the first step is A C1 C2 Ck , each Ci derives by a sequence < N steps. By the induction, each Ci is discovered by the algorithm to be nullable. So by the inductive step, A is eventually found to be nullable.

Eliminating -productions
If a grammar G1 is constructed by the elimination of -productions using the previous method of grammar G, then

L(G1) = L(G) - {}

Eliminating unit productions

The last part concerns the eliminating of unit productions Any production of the form A B , where A and B are variables, is called a unit production. These production introduce extra steps in the derivations that obviously are not needed in there.

Eliminating unit productions

Basis: (A, A) is a unit pair of any variable A, if A >* A by 0 steps. Induction: Lets (A, B) be a unit pair, and let B C is a production, where A, B, and C are variables, then we can conclude that (A, C) is also a unit pair. Theorem: The previous algorithm (basis and induction) finds exactly all the unit pairs for any grammar G.

Eliminating unit productions

Example 7.12

Eliminating unit productions

Example 7.12

Eliminating unit productions

Example 7.12
After eliminating the unit productions, the generated grammar is:

This grammar has no unit productions and still generates the same expressions as the previous one.

Chomsky Normal Form

Conclusion of all three elimination stages: Theorem: If G is a CFG which generates a language that consists of at least one string along with , then there is another CFG G1 such that: L{G1} = L{G} {} , no -productions, and G1 has neither unit productions nor useless symbols

Chomsky Normal Form

Proof: Start by performing the elimination of -productions. Then perform the elimination of unit productions, so the resulting grammar wont introduce any -productions since the new bodies are still identical to some bodies of the old grammar. Finally, perform the elimination of useless symbols, and since this eliminates productions and symbols, it will never reintroduce any -productions nor unit productions

Chomsky Normal Form

Every nonempty CFL without has grammar G in which all productions are in one of the following forms: A BC , where A, B, and C are variables or A a , where A is a variable and a is a terminal Also G doesnt contain any useless symbols A grammar complying to these forms is called a Chomsky Normal Form (CNF).

Chomsky Normal Form

1.

The construction of CNF is performed through: Arrangement of all bodies of length 2 or more to contain only variables. Breaking bodies of length 3 or more into a cascade productions, where each one has a body consisting of 2 variables.

2.

Chomsky Normal Form

Example 7.15

Chomsky Normal Form

Example 7.15 First: we introduce new variables to represent terminals:

Chomsky Normal Form

Example 7.15 Second: We make all bodies either a single terminal or multiple variables:

Chomsky Normal Form

Example 7.15 Last step: we make all bodies either a single terminal or two variables: