Bloom’s
CO# CO Statement
Level (L#)
CO Apply linguistic analysis methods to process and structure
1 textual data for better understanding of language L3
components
CO Use syntactic parsing techniques and data-driven models to
L4
2 analyze grammatical relationships within sentences.
CO Implement semantic interpretation and embedding models to
L3
3 capture meaning and contextual information in text.
CO Utilize statistical and machine learning–based language models to
L4
4 build adaptable and multilingual text processing systems.
CO Develop practical solutions for automated summarization and
L3
5 question answering using computational language techniques.
Module I
Finding the Structure of Words and Documents: Words and Their Components, Issues and Challenges,
Morphological Models. Introduction, Sentence Boundary Detection, Topic Boundary Detection,
Methods, Complexity of the Approaches, Performances of the Approaches , Features, Processing Stages.
Topic 1: Words and Their Components / Morphological Models
1. Words and Their Components
Short Questions (1M)
Understand (L2)
1. What is a morpheme?
2. Define a root word.
3. What is the difference between a prefix and a suffix?
Apply (L3)
4. Identify the morphemes in the word “unbelievable.”
5. Apply morphological segmentation to split “irresponsibility.”
Long Questions (5M)
Understand (L2)
1. Explain the components of words with suitable examples.
2. Discuss the importance of morphological structure in linguistic analysis.
Apply (L3) – Problem-Based
3. Apply morphological analysis to segment the following words and explain: internationalism,
miscommunication, overgeneralization.
4. Analyze a paragraph and identify root words, prefixes, and suffixes.
2. Issues and Challenges in Morphology
Short Questions (1M)
Understand (L2)
1. What is morphological ambiguity?
2. State one challenge in segmenting compound words.
Apply (L3)
3. Identify the ambiguity in the word “unlockable.”
4. Apply morphological analysis to explain why “record” has two different
pronunciations/meanings.
Long Questions (5M)
Understand (L2)
1. Explain issues faced in processing morphological variations.
2. Describe the types of morphological ambiguity with examples.
Apply (L3)
3. Apply techniques to resolve ambiguity in words like “rewrite,” “overhang,” “mislead.”
4. Apply morphological rules to address challenges in analyzing domain-specific terminology.
3. Morphological Models
Short Questions (1M)
Understand (L2)
1. What is a rule-based morphological model?
2. What is a lexicon in morphological processing?
Apply (L3)
3. Use a rule-based approach to segment “misinterpreted.”
4. Identify the stem of “nationalities” using a model.
Long Questions (5M)
Understand (L2)
1. Describe various morphological models used in NLP.
2. Explain the strengths and weaknesses of rule-based morphology.
Apply (L3)
3. Apply two morphological models (rule-based & statistical) to analyze a set of words and
compare outputs.
4. Apply a morphological model to extract features from a sample text.
4. Sentence Boundary Detection (SBD)
Short Questions (1M)
Understand (L2)
1. What is sentence boundary detection?
2. Why is SBD important?
Apply (L3)
3. Apply SBD rules to segment:
“Dr. Rao arrived at 5 p.m. He spoke at the seminar.”
4. Identify sentence boundaries in text containing abbreviations.
Long Questions (5M)
Understand (L2)
1. Explain rule-based and machine learning approaches for SBD.
2. Describe the challenges in detecting sentence boundaries.
Apply (L3)
3. Apply SBD to a noisy paragraph and show segmented output.
4. Apply ML-based SBD on a dataset and interpret boundary prediction results.
5. Topic Boundary Detection (TBD)
Short Questions (1M)
Understand (L2)
1. Define topic boundary detection.
2. What is a topic shift?
Apply (L3)
3. Identify the topic boundary in:
“Climate change is affecting weather patterns. Football fans gathered for the final match.”
4. Apply lexical cue analysis to mark a topic transition in a short passage.
Long Questions (5M)
Understand (L2)
1. Explain the need for topic boundary detection in text processing.
2. Describe similarity-based and ML-based TBD methods.
Apply (L3)
3. Apply TF–IDF or cosine similarity to detect a topic boundary in a multi-paragraph document.
4. Apply a TBD method to a news article and justify the detected boundaries.
6. Methods, Complexity & Performance
Short Questions (1M)
Understand (L2)
1. What is the complexity of rule-based SBD methods?
2. What is an evaluation metric for boundary detection?
Apply (L3)
3. Apply precision/recall to evaluate a sample SBD result.
4. Identify a factor that increases computational cost in TBD.
Long Questions (5M)
Understand (L2)
1. Discuss complexity issues in SBD and TBD.
2. Describe evaluation metrics for measuring system performance.
Apply (L3)
3. Apply evaluation metrics to compare two TBD systems using sample outputs.
4. Apply performance analysis to determine which SBD method is best for large documents.
7. Features & Processing Stages
Short Questions (1M)
Understand (L2)
1. What is a linguistic feature?
2. List the stages of text preprocessing.
Apply (L3)
3. Apply tokenization to segment a sentence.
4. Identify features useful for topic segmentation.
Long Questions (5M)
Understand (L2)
1. Explain the various stages involved in processing text for linguistic analysis.
Apply (L3)
2. Apply full processing stages (tokenization → normalization → feature extraction → boundary
detection) to a paragraph and present your output.
3. Apply feature analysis to detect structural changes in a document.
Module II
1. Parsing Natural Language
Short Questions (1M)
Apply (L3)
1. Apply syntactic parsing to identify the verb phrase in the sentence: “The child quickly finished
the homework.”
2. Apply phrase-structure rules to find the NP in: “A tall man in a blue coat entered.”
Analyze (L4)
3. Analyze why dependency parsing may produce different results for the word “run” in the
sentences: “I run daily.” and “I run a company.”
4. Analyze the syntactic role of “while” in: “She cooked while he cleaned.”
Long Questions (5M)
Apply (L3)
1. Apply constituent parsing to a given paragraph and draw the parse trees for any two sentences.
2. Apply syntactic parsing rules to identify all phrase types (NP, VP, PP, ADJP) in 3 selected
sentences.
Analyze (L4)
3. Analyze the structural differences between constituency and dependency parsing using
examples.
4. Analyze how parsing errors propagate in downstream NLP tasks (e.g., machine translation, QA).
2. Treebanks – A Data-Driven Approach to Syntax
Short Questions (1M)
Apply (L3)
1. Apply a Treebank annotation tag to classify the phrase: “in the room.”
2. Identify the POS tag sequence using Penn Treebank tags for: “The birds are flying.”
Analyze (L4)
3. Analyze why treebank annotations improve syntactic parsing accuracy.
4. Examine the difference between gold-standard and automatically generated treebanks.
Long Questions (5M)
Apply (L3)
1. Apply annotation guidelines to mark NP and VP structures in a small text segment.
2. Apply Treebank data to build a simple grammar rule set for parsing.
Analyze (L4)
3. Analyze how different treebanks (Penn, Universal, Negra, etc.) affect the design of parsers.
4. Analyze the limitations of Treebank-trained models when applied to informal or noisy text.
3. Representation of Syntactic Structure
Short Questions (1M)
Apply (L3)
1. Construct a simple dependency relation for: “The boy broke the window.”
2. Apply phrase-structure notation to represent a noun phrase.
Analyze (L4)
3. Analyze the syntactic structure that allows movement in wh-questions like “What did you buy?”
4. Analyze whether PP-attachment is ambiguous in: “She saw the man with a telescope.”
Long Questions (5M)
Apply (L3)
1. Apply syntactic structure rules to draw phrase structure trees for 3 given sentences.
2. Convert constituency trees into dependency trees for selected examples.
Analyze (L4)
3. Analyze different syntactic representation models (dependency, CCG, HPSG) with examples.
4. Analyze how structural representation influences semantic interpretation in an NLP system.
4. Parsing Algorithms
Short Questions (1M)
Apply (L3)
1. Apply CYK parsing to check whether “the cat sleeps” fits a simple CFG.
2. Identify which parsing technique (top-down or bottom-up) suits a left-recursive grammar.
Analyze (L4)
3. Analyze why Earley parsing handles ambiguity better.
4. Analyze factors that affect the time complexity of dynamic programming parsers.
Long Questions (5M)
Apply (L3)
1. Apply a chosen parsing algorithm (Earley, CYK, shift-reduce) to parse a 3-line input text.
2. Apply bottom-up parsing to derive a valid parse tree for a CFG grammar given.
Analyze (L4)
3. Compare and analyze the efficiency of Earley vs. CYK parsing with example grammars.
4. Analyze how parser performance changes when grammar size increases significantly.
5. Models for Ambiguity Resolution in Parsing
Short Questions (1M)
Apply (L3)
1. Apply a disambiguation rule to resolve PP-attachment in: “He ate the cake with a spoon.”
2. Apply probabilistic parsing to choose the most likely parse of a given sentence.
Analyze (L4)
3. Analyze structural ambiguity in the sentence: “Old men and women sat together.”
4. Analyze why lexical ambiguity causes multiple valid parse trees.
Long Questions (5M)
Apply (L3)
1. Apply statistical methods to resolve ambiguity in syntactic parsing for 3 sentences.
2. Apply rule-based methods to resolve syntactic ambiguity in a given paragraph.
Analyze (L4)
3. Analyze different ambiguity types (lexical, structural, scope) and how parsers address them.
4. Analyze ambiguity resolution performance using probabilistic vs. neural parsers.
6. Multilingual Issues in Syntax Parsing
Short Questions (1M)
Apply (L3)
1. Apply universal dependency notation to annotate a non-English sentence.
2. Identify a syntactic challenge when parsing free-word-order languages.
Analyze (L4)
3. Analyze the effects of morphological richness on parsing accuracy.
4. Analyze why a parser trained on English fails on Tamil/Hindi without adaptation.
Long Questions (5M)
Apply (L3)
1. Apply multilingual parsing techniques to annotate 2 sentences each in English and another
language.
2. Apply cross-lingual transfer to parse a sentence from a low-resource language.
Analyze (L4)
3. Analyze multilingual parsing challenges, including morphology, word order, and script variations.
4. Analyze the performance differences between monolingual, multilingual, and cross-lingual
parsers.