AL-504-B Natural Language Processing - Dec 2025 Question Paper

Under each question, How to answer lists the points an answer should cover, outlined by CampusPrep from the syllabus. Check the details against your textbook.

1. 7 Marks
   aDefine Natural Language Processing (NLP) and explain its complete historical evolution from rule-based systems to modern deep learning based NLP models. Discuss in detail the major challenges in natural language understanding, ambiguity resolution, context handling and multilingual processing with suitable real-world examples.
   How to answer

   • Definition of NLP as intersection of linguistics, CS and AI
   • Historical evolution: rule-based, statistical, machine learning, deep learning/transformers
   • Major challenges: ambiguity, context, multilinguality with examples
   • Real-world examples for each challenge
2. 7 Marks
   bExplain the levels of language analysis in NLP. Describe lexical, syntactic, semantic, discourse and pragmatic analysis with examples.
   How to answer

   • Five levels of linguistic analysis overview
   • Lexical, syntactic, semantic, discourse, pragmatic with definition each
   • Example for each level
   • Role in NLP pipeline
3. 7 Marks
   aExplain tokenization, stemming and lemmatization in detail. Compare stemming and lemmatization and discuss their importance in NLP applications.
   How to answer

   • Definition of tokenization, stemming, lemmatization with examples
   • Stemming vs lemmatization comparison table
   • Algorithms examples: Porter stemmer, WordNet lemmatizer
   • Importance in search, IR, classification applications
4. 7 Marks
   bDescribe stop word removal and normalization techniques. Explain how text pre-processing improves NLP system performance.
   How to answer

   • Stop word removal concept and lists
   • Normalization: case folding, punctuation, numbers handling
   • Text preprocessing pipeline steps
   • Impact on vocabulary size, accuracy and efficiency
5. 7 Marks
   aExplain Part-of-Speech (POS) tagging in detail. Describe different approaches used for POS tagging such as rule-based, stochastic and neural approaches.
   How to answer

   • Definition and purpose of POS tagging
   • Rule-based approach with examples
   • Stochastic approach: HMM, MEMM, CRF
   • Neural approaches: BiLSTM, transformers
6. 7 Marks
   bDiscuss ambiguity in natural language processing. Explain lexical ambiguity and syntactic ambiguity with suitable examples.
   How to answer

   • Concept of ambiguity in NLP
   • Lexical ambiguity: homonymy, polysemy with examples
   • Syntactic ambiguity: PP attachment with examples
   • Need for disambiguation
7. 7 Marks
   aExplain syntax analysis in NLP. Describe top-down and bottom-up parsing techniques with suitable examples.
   How to answer

   • Definition of syntax analysis and parsing
   • Top-down parsing: start from S with example
   • Bottom-up parsing: shift-reduce with example
   • Comparison and limitations
8. 7 Marks
   bExplain Context Free Grammar (CFG) and parse trees. Discuss their role in syntactic analysis.
   How to answer

   • CFG definition: terminals, non-terminals, productions
   • Example grammar and parse tree construction
   • Role in syntactic analysis and grammar checking
   • Limitations and extensions
9. 7 Marks
   aExplain semantic analysis in Natural Language Processing. Discuss meaning representation techniques, semantic roles, thematic relations and logical forms. Explain how semantic analysis helps in tasks such as question answering, information extraction and machine translation.
   How to answer

   • Definition of semantic analysis
   • Meaning representation, semantic roles, thematic relations, logical forms
   • Examples of agent, patient, instrument roles
   • Use in QA, IE and machine translation
10. 7 Marks
    bDiscuss Word Sense Disambiguation (WSD). Explain different approaches used for WSD with examples.
    How to answer

    • Definition of Word Sense Disambiguation
    • Knowledge-based, supervised, unsupervised approaches
    • Examples: Lesk algorithm, classifiers
    • Applications and challenges
11. 7 Marks
    aExplain language models in NLP. Describe N-gram language models and discuss their advantages and limitations.
    How to answer

    • Definition of language models
    • N-gram concept: unigram, bigram, trigram with probability formula
    • Advantages: simplicity, efficiency
    • Limitations: sparsity, long-range dependency
12. 7 Marks
    bExplain neural language models. Compare traditional statistical language models with neural language models.
    How to answer

    • Neural language models: embeddings, RNN, LSTM, transformers
    • Statistical vs neural comparison on context and generalization
    • Advantages of neural models
    • Limitations: data and compute requirement
13. 7 Marks
    aExplain Machine Translation in Natural Language Processing. Describe rule-based statistical and neural machine translation approached in detail. Compare these approaches in terms of accuracy, scalability and real-world applications.
    How to answer

    • Definition of machine translation
    • Rule-based, statistical, neural MT working
    • Comparison on accuracy, scalability
    • Real-world applications like Google Translate
14. 7 Marks
    bDiscuss major applications of NLP in real-world systems such as information retrieval, sentiment analysis and chatbots.
    How to answer

    • Overview of NLP applications
    • Information retrieval working
    • Sentiment analysis techniques
    • Chatbots architecture and examples
15. 7 Marks
    aExplain Speech Recognition systems and their working principles.
    How to answer

    • Definition of speech recognition
    • Components: acoustic model, language model
    • Working pipeline: feature extraction to decoding
    • Challenges and uses
16. 7 Marks
    bExplain Text Summarization techniques and their applications.
    How to answer

    • Definition of text summarization
    • Extractive vs abstractive techniques
    • Algorithms: TF-IDF, TextRank, seq2seq
    • Applications in news, legal, research