Academy / Data & Machine Learning

Path 2 · Course 4 · For analysts and aspiring practitioners

NLP Fundamentals

Turn text into insight with Python: tokens and bag of words, TF-IDF classifiers, embeddings and semantic search, information extraction with regex and LLMs, transformers and few-shot classification, topic discovery, and responsible NLP in production.

📘 6 modules⏱ About 4 hours✅ Knowledge check per module🎓 Final assessment & certificate🎯 Python, pandas & basic ML

0 of 6 modules complete

Start Module 1 Final assessment

Modules

  1. 1

    Text as data

    What NLP is used for, the practice dataset, tokenisation and normalisation, bag of words, stop words and n-grams.

    30 min
  2. 2

    Text classification

    TF-IDF, a vectoriser-plus-model pipeline, inspecting learned words, error analysis and sentiment.

    35 min
  3. 3

    Embeddings and semantic search

    Why exact word matching falls short, cosine similarity, getting embeddings from an API, semantic search steps and the limits of embeddings.

    35 min
  4. 4

    Information extraction

    Regular expressions for precise patterns, named entity recognition, LLM extraction to JSON, validation and grounding checks.

    35 min
  5. 5

    Transformers and LLMs for NLP

    From counts to transformers, attention, tokens and cost, zero- and few-shot classification, and choosing between classic NLP and LLMs.

    35 min
  6. 6

    NLP in practice

    Topic modelling with NMF, tracking language over time, personal data and redaction, bias across languages and dialects, and running NLP in production.

    35 min