Natural Language Processing (AL-504 (B)) - Important Questions

Define Natural Language Processing and explain its key components and challenges.

Explain tokenization, stemming and lemmatization, compare stemming vs lemmatization, and their importance in morphological analysis for NLP.

Define formal grammar, explain its key components, and discuss its role in language modeling for NLP.

Compare common types of smoothing techniques used in NLP and how they differ.

Compare Hidden Markov Models and Maximum Entropy models in terms of modeling approach, key components and probabilistic dependencies.

Compute unsmoothed unigram and bigram probabilities for a given training corpus and calculate the probability of a test sentence using N-grams. Discuss the problem of zero probabilities.

Explain how dependency grammar differs from phrase-structure grammar.

Explain syntax analysis / syntactic parsing in NLP, including top-down and bottom-up parsing techniques with suitable examples.

Check acceptance of a given string using the CYK algorithm for the grammar S -> AB / BC, A -> BA / a, B -> CC / b, C -> AB / a.

Explain ambiguity in NLP including lexical and syntactic / parse-tree ambiguity with examples.

Illustrate word similarity using a thesaurus and distributional methods.

Explain supervised Word Sense Disambiguation methods with suitable examples.

Briefly explain different bootstrapping methods in NLP.

Describe the functioning of machine translation as an intelligent work processor with real-world application examples. Explain rule-based, statistical and neural approaches.

Explain major commercial and real-world applications of NLP and how they improve user experience and interaction.

Explain speech recognition systems, their working principles, and how NLP enhances this technology.