Learning path
NLP: Working with Text
Cleaning and tokenising, TF-IDF and similarity, sentiment analysis and transformers, with animated films and runnable Python. Part of the Data Scientist premium track.
-
1
NLP Basics: Cleaning and Tokenising Text
Prepare text for machine learning: lowercasing, removing noise, tokenisation, stop words, stemming vs lemmatisation, n-grams, and handling Hinglish and emojis, with runnable Python.
-
2
TF-IDF, Text Similarity and Search
Turn documents into vectors with bag-of-words and TF-IDF, compare them with cosine similarity, build a mini search engine, and measure retrieval with precision and recall. With scikit-learn.
-
3
Sentiment Analysis: From TF-IDF to Transformers
Classify reviews as positive or negative with a TF-IDF + logistic regression baseline you can run, then with a pretrained transformer from Hugging Face, and know when to…