Data Science Natural Language Processing (NLP) Syllabus

Natural Language Processing (NLP) syllabus — full chapter and topic list

Expand any chapter to see its topics and sub-topics. This is the whole examinable outline for Natural Language Processing (NLP) in Data Science, not a summary of it.

1. 1

   Introduction to Natural Language Processing (NLP)

   3 topics

   • What is NLP?
     • Definition and Scope
     • History and Evolution
   • Applications of NLP
     • Text Classification
     • Sentiment Analysis
     • Machine Translation
     • Chatbots and Conversational Agents
     • Speech Recognition
   • Challenges in NLP
     • Ambiguity
     • Context and Pragmatics
     • Resource Scarcity
2. 2

   Basic Linguistics for NLP

   3 topics

   • Syntax
     • Parts of Speech
     • Phrase Structure
   • Semantics
     • Meaning Representation
     • Lexical Semantics
   • Pragmatics
     • Contextual Meaning
     • Speech Acts
3. 3

   Text Preprocessing

   3 topics

   • Tokenization
     • Word Tokenization
     • Sentence Tokenization
   • Normalization
     • Lowercasing
     • Stemming
     • Lemmatization
   • Removing Noise
     • Stop Words
     • Punctuation
     • Special Characters
4. 4

   Feature Extraction

   3 topics

   • Bag of Words (BoW)
     • Term Frequency
     • Inverse Document Frequency (IDF)
   • TF-IDF
     • Calculation
     • Applications
   • Word Embeddings
     • Word2Vec
     • GloVe
     • FastText
5. 5

   NLP Algorithms

   3 topics

   • Text Classification
     • Naive Bayes
     • Support Vector Machines (SVM)
     • Neural Networks
   • Sequence Models
     • Hidden Markov Models (HMMs)
     • Conditional Random Fields (CRFs)
   • Neural Networks for NLP
     • Recurrent Neural Networks (RNNs)
     • Long Short-Term Memory (LSTM)
     • Gated Recurrent Units (GRUs)
6. 6

   Advanced NLP Techniques

   3 topics

   • Transformers
     • Attention Mechanism
     • BERT
     • GPT
   • Transfer Learning in NLP
     • Pre-trained Models
     • Fine-Tuning
   • Evaluation Metrics
     • Precision, Recall, F1-Score
     • BLEU Score
     • ROUGE Score
7. 7

   NLP in Practice

   3 topics

   • NLP Libraries and Frameworks
     • NLTK
     • spaCy
     • Hugging Face Transformers
   • Building NLP Pipelines
     • Data Collection
     • Preprocessing
     • Model Training
     • Evaluation and Deployment
   • Case Studies
     • Sentiment Analysis in Social Media
     • Machine Translation at Scale
     • Chatbots for Customer Service

Natural Language Processing (NLP) flashcards for Data Science

20 of 51 cards from the Natural Language Processing (NLP) deck — real questions with worked answers.

1. What is Natural Language Processing (NLP)?

   NLP is a subfield of artificial intelligence and linguistics concerned with enabling computers to understand, interpret, generate, and respond to human (natural) language in text or speech form.

2. What are the two broad subfields that NLP combines techniques from?

   NLP combines computational linguistics (rule-based modeling of language) with machine learning, deep learning, and statistical models.

3. Distinguish Natural Language Understanding (NLU) from Natural Language Generation (NLG).

   NLU focuses on machine reading comprehension — mapping language to meaning (intent, entities). NLG focuses on producing fluent, coherent natural language text from structured data or internal representations.

4. List four common real-world applications of NLP.

   Machine translation, sentiment analysis, chatbots/virtual assistants, spam detection, text summarization, named-entity recognition, speech recognition, and autocomplete/autocorrect.

5. What NLP application assigns a polarity (positive, negative, neutral) to a piece of text?

   Sentiment analysis (opinion mining).

6. Name three major challenges that make NLP difficult.

   Ambiguity (lexical, syntactic, semantic), context dependence, sarcasm/irony, idioms and figurative language, spelling/grammatical errors, language diversity, and coreference resolution.

7. What is lexical ambiguity? Give the type of word it involves.

   Lexical ambiguity occurs when a single word has multiple meanings (polysemy/homonymy), e.g. 'bank' meaning a riverbank or a financial institution.

8. What is syntactic (structural) ambiguity?

   When a sentence can be parsed into more than one grammatical structure, yielding different meanings, e.g. 'I saw the man with the telescope.'

9. In linguistics/NLP, what does syntax refer to?

   Syntax is the set of rules governing the grammatical arrangement of words and phrases to form valid sentences — sentence structure, not meaning.

10. What is semantics in NLP?

    Semantics is the study of meaning — the literal meaning of words, phrases, and sentences independent of context.

11. What is pragmatics in NLP?

    Pragmatics is the study of how context, speaker intent, and real-world knowledge affect the interpretation of language beyond its literal meaning.

12. Order these three from form to context: semantics, pragmatics, syntax.

    Syntax (structure/form) → Semantics (literal meaning) → Pragmatics (contextual/intended meaning).

13. What is tokenization in NLP?

    Tokenization is the process of breaking text into smaller units called tokens — typically words, subwords, or characters — for further processing.

14. Differentiate word tokenization from subword tokenization.

    Word tokenization splits text on word boundaries (whitespace/punctuation). Subword tokenization (e.g. Byte-Pair Encoding, WordPiece) splits words into smaller frequent units to handle rare and out-of-vocabulary words.

15. What is text normalization?

    Normalization transforms text into a single canonical form, e.g. lowercasing, expanding contractions, converting numbers/dates to standard formats, and reducing words to base forms.

16. Compare stemming and lemmatization.

    Stemming crudely chops word endings to a root using heuristics (e.g. 'studies' → 'studi'), possibly non-words. Lemmatization uses vocabulary and morphology to return the valid dictionary base form/lemma (e.g. 'studies' → 'study').

17. What are stop words and why are they often removed?

    Stop words are very common words (the, is, at, and) that carry little discriminative meaning; removing them reduces dimensionality and noise in many bag-of-words style models.

18. What is meant by 'removing noise' in text preprocessing?

    Eliminating irrelevant or distracting elements such as HTML tags, punctuation, special characters, URLs, extra whitespace, and emojis that do not contribute to the analysis.

19. What is the Bag of Words (BoW) model?

    BoW represents a document as an unordered multiset of its words, recording word counts (or presence) while discarding grammar and word order.

20. State one key limitation of the Bag of Words model.

    BoW ignores word order and context, produces high-dimensional sparse vectors, and treats all words as independent, so it cannot capture semantic similarity between words.

See more Natural Language Processing (NLP) flashcards →

Planning Natural Language Processing (NLP) for Data Science

Natural Language Processing (NLP) (Data Science) FAQ