Lexical, Syntactic and Semantic processing Interview Questions

-

 

🧠 Lexical Processing (Q1–Q17)

Lexical processing focuses on individual words and their forms.

🔹 Questions & Answers

  1. What is lexical analysis in NLP?
    → It breaks text into tokens (words, punctuation) for further processing.

  2. Define tokenization.
    → Splitting text into meaningful units like words or sentences.

  3. What is stemming?
    → Reducing words to their root form (e.g., “running” → “run”).

  4. What is lemmatization?
    → Mapping words to their dictionary root using linguistic rules.

  5. Difference between stemming and lemmatization?
    → Stemming is rule-based and faster; lemmatization is more accurate and context-aware.

  6. What are stop words?
    → Common words (e.g., “the”, “is”) removed to reduce noise.

  7. What is a corpus?
    → A large collection of text used for training NLP models.

  8. What is a vocabulary in NLP?
    → The set of unique tokens in a corpus.

  9. What is n-gram modeling?
    → A sequence of n words used to capture context (e.g., bigrams, trigrams).

  10. What is frequency distribution?
    → A count of how often each token appears.

  11. What is Zipf’s Law?
    → In natural language, the frequency of any word is inversely proportional to its rank.

  12. What is TF-IDF?
    → Term Frequency–Inverse Document Frequency, used to weigh word importance.

  13. What is word normalization?
    → Converting words to a standard format (e.g., lowercase, removing punctuation).

  14. What is text cleaning?
    → Removing noise like HTML tags, emojis, or special characters.

  15. What is a tokenizer in spaCy or NLTK?
    → A tool that splits text into tokens using language-specific rules.

  16. What is subword tokenization?
    → Breaking words into smaller units (e.g., Byte Pair Encoding).

  17. What is out-of-vocabulary (OOV)?
    → Words not present in the training vocabulary.

🧩 Syntactic Processing (Q18–Q33)

Syntactic processing deals with grammar and sentence structure.

🔹 Questions & Answers

  1. What is syntactic analysis?
    → Parsing sentences to understand grammatical structure.

  2. What is POS tagging?
    → Assigning part-of-speech labels (noun, verb, etc.) to words.

  3. What is a parse tree?
    → A tree structure representing sentence grammar.

  4. What is dependency parsing?
    → Identifying relationships between words (e.g., subject → verb).

  5. What is constituency parsing?
    → Dividing sentences into sub-phrases like noun phrases (NP) and verb phrases (VP).

  6. What is a context-free grammar (CFG)?
    → A set of rules defining valid sentence structures.

  7. What is chunking?
    → Grouping tokens into meaningful phrases (e.g., noun chunks).

  8. What is shallow parsing?
    → Identifying phrases without full syntactic structure.

  9. What is POS ambiguity?
    → When a word can have multiple POS tags depending on context.

  10. What is a syntax error in NLP?
    → A violation of grammatical rules in a sentence.

  11. What is a universal POS tagset?
    → A standardized set of POS tags used across languages.

  12. What is morphological analysis?
    → Studying word structure (prefixes, suffixes, roots).

  13. What is a language model?
    → A model that predicts the next word in a sequence.

  14. What is the role of syntax in machine translation?
    → Ensures grammatical correctness in translated output.

  15. What is syntactic ambiguity?
    → When a sentence has multiple valid grammatical interpretations.

  16. What is a parser?
    → A tool that builds syntactic structure from text.

🧠 Semantic Processing (Q34–Q50)

Semantic processing extracts meaning from text.

🔹 Questions & Answers

  1. What is semantic analysis?
    → Understanding the meaning of words, phrases, and sentences.

  2. What is Named Entity Recognition (NER)?
    → Identifying entities like names, dates, and locations.

  3. What is word sense disambiguation (WSD)?
    → Determining the correct meaning of a word in context.

  4. What is semantic similarity?
    → Measuring how similar two texts are in meaning.

  5. What is sentiment analysis?
    → Classifying text as positive, negative, or neutral.

  6. What is semantic role labeling (SRL)?
    → Identifying roles like agent, action, and object in a sentence.

  7. What is co-reference resolution?
    → Linking pronouns to the entities they refer to.

  8. What is entailment in NLP?
    → Determining if one sentence logically follows from another.

  9. What is a knowledge graph?
    → A structured representation of entities and their relationships.

  10. What is a semantic parser?
    → Converts natural language into machine-readable meaning representations.

  11. What is polysemy?
    → A word with multiple related meanings (e.g., “bank”).

  12. What is homonymy?
    → A word with multiple unrelated meanings (e.g., “bat”).

  13. What is lexical semantics?
    → Study of word meanings and relationships.

  14. What is distributional semantics?
    → Inferring meaning from word usage patterns in context.

  15. What is embedding in NLP?
    → Vector representation of words capturing semantic meaning.

  16. What is contextual embedding?
    → Word vectors that change based on surrounding context (e.g., BERT).

  17. What is semantic drift?
    → Change in word meaning over time or across domains.





🧠 Lexical Processing (Word-Level Analysis)

Lexical processing deals with individual words—their structure, meaning, and role in a sentence.

🔤 Questions & Answers

  1. What is lexical analysis in NLP?
    It involves breaking down text into tokens (words, punctuation, etc.) and identifying their lexical categories.

  2. What is tokenization?
    It's the process of splitting text into individual units like words or sentences.

  3. What is stemming?
    Reducing words to their root form (e.g., "running" → "run").

  4. What is lemmatization?
    Similar to stemming but uses vocabulary and morphological analysis to return the base form (e.g., "better" → "good").

  5. Difference between stemming and lemmatization?
    Stemming is rule-based and faster; lemmatization is dictionary-based and more accurate.

  6. What is a lexicon in NLP?
    A collection of words and their meanings or properties.

  7. What is part-of-speech tagging?
    Assigning grammatical categories (noun, verb, etc.) to each word.

  8. How does POS tagging help in NLP?
    It aids in syntactic parsing, named entity recognition, and semantic analysis.

  9. What are stop words?
    Common words (like "the", "is") that are often removed during preprocessing.

  10. What is word segmentation?
    Dividing text into meaningful words, especially in languages without spaces (e.g., Chinese).

🧩 Syntactic Processing (Grammar & Structure)

Syntactic processing focuses on sentence structure and grammatical relationships.

🧠 Questions & Answers

  1. What is syntactic analysis?
    It checks if the sentence structure conforms to grammar rules.

  2. What is parsing in NLP?
    Analyzing sentence structure to determine relationships between words.

  3. What is a parse tree?
    A tree diagram representing the syntactic structure of a sentence.

  4. What is dependency parsing?
    It identifies relationships between words (e.g., subject → verb).

  5. What is constituency parsing?
    It breaks a sentence into sub-phrases (noun phrase, verb phrase, etc.).

  6. What is POS ambiguity?
    When a word can belong to multiple parts of speech (e.g., "book" as noun or verb).

  7. What is syntax in NLP?
    Rules that govern sentence structure and word order.

  8. How does syntactic processing differ from semantic processing?
    Syntax is about structure; semantics is about meaning.

  9. What is chunking in NLP?
    Grouping words into meaningful phrases (e.g., noun phrases).

  10. What are CFGs (Context-Free Grammars)?
    Formal grammar rules used to define syntactic structure.

🧠 Semantic Processing (Meaning & Context)

Semantic processing aims to understand the meaning of text.

🧠 Questions & Answers

  1. What is semantic analysis in NLP?
    Understanding the meaning of words, phrases, and sentences.

  2. What is named entity recognition (NER)?
    Identifying proper nouns like names, places, and organizations.

  3. What is word sense disambiguation?
    Determining the correct meaning of a word based on context.

  4. What is semantic role labeling?
    Identifying the role played by words in a sentence (e.g., agent, action, object).

  5. What is co-reference resolution?
    Linking pronouns and noun phrases to the same entity.

  6. What is sentiment analysis?
    Determining the emotional tone of text.

  7. What is a semantic network?
    A graph representing relationships between concepts.

  8. What is distributional semantics?
    Inferring word meaning based on surrounding words.

  9. What is a knowledge graph?
    A structured representation of entities and their relationships.

  10. What is semantic similarity?
    Measuring how similar two pieces of text are in meaning.

🧪 Advanced & Applied Questions

  1. How do embeddings help in semantic processing?
    They convert words into vectors capturing semantic relationships.

  2. What is the role of transformers in semantic analysis?
    They capture long-range dependencies and contextual meaning.

  3. How does BERT improve semantic understanding?
    BERT uses bidirectional context to better understand word meaning.

  4. What is the difference between lexical and semantic ambiguity?
    Lexical ambiguity is word-level; semantic ambiguity is sentence-level.

  5. How is semantic parsing different from syntactic parsing?
    Semantic parsing maps text to meaning representations; syntactic parsing maps to grammar trees.

  6. What is entailment in NLP?
    Determining if one sentence logically follows from another.

  7. What is pragmatic analysis?
    Understanding meaning based on context and speaker intent.

  8. What is discourse analysis?
    Studying language beyond individual sentences—e.g., conversation flow.

  9. What is the role of ontologies in semantic processing?
    They define concepts and relationships for domain-specific understanding.

  10. How does semantic search differ from keyword search?
    Semantic search understands intent and meaning; keyword search matches exact terms.

🧠 Bonus: Interview-Style Conceptual Questions

  1. Explain the NLP pipeline from lexical to semantic processing.
    Tokenization → POS tagging → Parsing → Semantic analysis → Output.

  2. How do you handle ambiguity in NLP?
    Use context-aware models like BERT or apply disambiguation algorithms.

  3. What are challenges in semantic processing?
    Ambiguity, sarcasm, idioms, and context sensitivity.

  4. How does semantic processing help in chatbots?
    It enables understanding of user intent and generating meaningful responses.

  5. What is the role of embeddings in semantic similarity?
    Embeddings capture contextual meaning, enabling similarity comparisons.

  6. How do you evaluate semantic models?
    Using metrics like BLEU, ROUGE, cosine similarity, and human evaluation.

  7. What is the difference between syntax-based and semantics-based models?
    Syntax-based models focus on structure; semantics-based models focus on meaning.

  8. How does semantic processing help in question answering systems?
    It enables understanding of both the question and relevant context.

  9. What is the role of semantic parsing in RAG pipelines?
    It helps convert natural language queries into structured formats for retrieval.

  10. How do you integrate lexical, syntactic, and semantic layers in a real-world NLP system?
    Through a modular pipeline or transformer-based models that encode all layers jointly.

Comments

Post a Comment

Popular posts from this blog

Resume Work and Project Details

-
 Conceptual / MMM Basics Q1. What is Media Mix Modeling (MMM)? MMM quantifies how different marketing channels and external factors drive business outcomes such as sales or conversions, using regression-based or Bayesian models on historical time-series data. Unlike attribution models, MMM doesn’t require user-level data and works even with aggregated spend data. Q2. What problem does MMM solve? It helps marketers measure ROI by channel and optimize future spend allocation while accounting for carryover, seasonality, and external effects. Q3. What is the dependent variable and what are the independent variables in your models? Dependent variable: Sales or revenue. Independent variables: Marketing touchpoints (TV, Digital, Print, Radio) and control variables (Price, Competitor, Macro factors, Seasonality). Q4. How is MMM different from digital attribution models? Attribution models rely on user-level tracking (cookies, clicks). MMM uses aggregate, privacy-compliant data and...
Read more

Time Series and MMM basics

-
🕒 1. What is Time Series Data? Definition: Time series data is a sequence of observations recorded at specific time intervals (daily, weekly, monthly, etc.). Each data point has two key components: Time (t): a timestamp (e.g., date, hour, quarter) Value (y): a measurement (e.g., sales, ad spend, temperature) In Media Mix Modeling: time series data might include weekly sales, ad impressions, or marketing spend across channels. Example: Week TV Spend Online Spend Sales 1 1000 500 15000 2 1200 600 16200 3 900 550 14800 4 1100 700 17000 Here, Sales is the time series variable we want to analyze or forecast. 📊 2. What is Time Series Analysis? Definition: Time series analysis is the study of data points collected over time to identify patterns , trends , seasonality , and dependencies —and to model or forecast future behavior. In MMM, this helps us understand how marketing efforts (TV, digital, print) influence sales over time and fore...
Read more

LINEAR REGRESSION

-
Image
Machine Learning Machine Learning (ML) is a branch of artificial intelligence that focuses on building systems that learn from and make decisions based on data. There are several types of ML, but let’s focus on the two main categories:  Supervised  and  Unsupervised Learning . Supervised Learning Definition : In supervised learning, the algorithm is trained on labeled data. This means the input data is paired with the correct output, and the goal is to learn a mapping from inputs to outputs. Examples : Classification : Predicting whether an email is spam or not spam. Regress...
Read more