Bag of Words & TF-IDF

Summary

Classic text representation methods — Bag of Words counts word occurrences, TF-IDF weights words by how distinctive they are to a document versus the whole corpus — simple, fast, and still useful baselines.

• 1Build a Bag of Words representation for a small set of sample documents.
• 2Compute TF-IDF scores and identify the most 'distinctive' words per document.
• 3Learn why TF-IDF often outperforms raw word counts for text classification.