(SEM V) THEORY EXAMINATION 2024-25 NATURAL LANGUAGE PROCESSING

Course: B.Tech (Semester V)

Subject: Natural Language Processing

Subject Code: BCAI052

Maximum Marks: 70

Time: 3 Hours

Paper ID: 310316

Section A (2 × 7 = 14 Marks)

Answer all questions briefly:

Write various limitations of Natural Language Processing.

Differentiate between top-down and bottom-up parsers.

Discuss the concept of Knowledge Representation and Knowledge Base.

What are the limitations of Probabilistic Context-Free Grammars (PCFGs)?

Write the concept of Phonetics and list its main branches.

Discuss the applications of filter bank methods in speech signal processing.

What does discourse planning involve? 

Section B (Attempt any three × 7 = 21 Marks)

Explain Add-k smoothing and discuss methods for evaluating N-grams.

Provide an algorithm for a basic top-down parser with an example.

Describe thesaurus-based and distributional word similarity methods with their advantages.

Explain the concept of Digital Signal Processing.

Discuss parallelization and batch processing techniques, and compare both in the context of NLP. 

Section C (Long Answer Questions – 7 Marks each)

Q3

(a) Provide a pseudo-code algorithm for parsing a Finite-State Transducer and explain with an example,
OR

(b) Explain Zipf’s Law and how it is applied in NLP.

Q4

(a) Explain Augmented Grammars and Lexicon in NLP,
OR

(b) For CFGs (S → NP VP, VP → V NP, NP → Det N), draw a Shift-Reduce Parser for the sentence:
“The woman saw a puppy.”
Lexical entries:

The, a → Det

Woman, puppy → N

Saw → V

Q5

(a) Explain the relationships between word senses and knowledge sources used in Word Sense Disambiguation (WSD),
OR

(b) Explain Thematic Roles in semantic analysis and their applications.

Q6

(a) Explain vowels, semi-vowels, diphthongs, and consonants in relation to Speech Recognition Systems,
OR

(b) Explain Linear Predictive Coding (LPC) — how it models speech signals and its benefits in speech analysis and synthesis.

Q7

(a) Explain the Hidden Markov Model (HMM) with Baum-Welch parameter re-estimation and its implementation issues,
OR

(b) Describe the concept and process of Chatbots with an example. 

Important Topics for Revision

Limitations & Scope of NLP

Parsing Techniques (Top-Down, Bottom-Up, Shift-Reduce)

Knowledge Representation, PCFG, and Lexicon

Word Sense Disambiguation (WSD)

Phonetics and Speech Signal Processing

Hidden Markov Models & Chatbots

Parallelization and Batch Processing in NLP

Digital Signal & Speech Processing Concepts