Chomsky Normal Form
Duration: 6 min
This video lesson is available to enrolled students.
AI Summary
An AI-generated summary of this video lecture.
The lecture introduces the concept of Normalization for Context-Free Grammars (CFG), specifically focusing on Chomsky Normal Form (CNF). The instructor defines CNF as a state where every production rule follows the format A → BC or A → a, where B and C are non-terminals and a is a terminal. The session transitions into a practical demonstration using the grammar S → aSb / ab. The instructor systematically converts this grammar into CNF by introducing auxiliary non-terminals to replace terminals and breaking down right-hand sides that exceed length two. Finally, the lecture concludes by deriving a mathematical property: for any string of length n derived from a CNF grammar, the number of production steps required is exactly 2n - 1.
Chapters
0:00 – 2:00 00:00-02:00
The video begins with a slide titled 'Normalization of CFG'. The text defines Chomsky Normal Form, stating that a grammar G is in CNF if every production is in the form A → BC / a. Below this, the constraints are listed: B, C ∈ Vn (non-terminals) and a ∈ Σ (terminals). The instructor underlines the phrase 'Chomsky Normal Form' and uses red arrows to point to the components A, B, C, and a on the slide, emphasizing the structural requirements of the rule. He specifically highlights that B and C must be non-terminals from the set Vn, while a must be a terminal from the set Σ.
2:00 – 5:00 02:00-05:00
The instructor moves to a whiteboard example with the production S → aSb / ab. He explains that this is not in CNF because the right-hand side contains terminals mixed with non-terminals and has a length of three. He introduces new non-terminals A and B with productions A → a and B → b. He rewrites the main rule as S → ASB / AB. Since ASB has length three, he introduces a new variable S1 (written as SA or similar) to split it, creating S1 → AS. The final normalized grammar is listed as S → S1 B / AB, S1 → AS, A → a, and B → b. He emphasizes that every rule must now strictly follow the A → BC or A → a format.
5:00 – 6:29 05:00-06:29
The instructor discusses the relationship between string length and derivation steps. He writes |w| = n and states that for a CNF grammar, the number of productions needed is 2n - 1. He demonstrates this with a calculation: 2(4) - 1 = 7. He lists the sequence of sentential forms for a derivation of length 4: S, S1 B, ASB, aSB, aAB, aBb, ab. This confirms that there are exactly 7 steps, matching the formula. He circles the final string ab to show the result. The slide text reiterates that if a CFG is in CNF, the number of sentential forms for a string of length n is 2n - 1.
The lesson effectively bridges theory and practice. It begins by defining the rigid structural constraints of Chomsky Normal Form, ensuring productions are binary or unary. It then applies these constraints to normalize a specific grammar, demonstrating the necessary introduction of new variables to handle mixed terminals and long sequences. The session concludes by establishing a critical algorithmic property: the linear relationship between string length and derivation steps (2n - 1), which is fundamental for parsing algorithms like CYK.