TF-IDF
Domain: NLP Fundamentals
π― Motivation
Term Frequency x Inverse Document Frequency. Highlights words that are important in a document but rare overall.
π Concrete Example
The appears in every document (low IDF, ignored). Quantum might appear in one physics paper (high TF-IDF, important).
π Analogy
Like a wordβs VIP score - frequent in this document but rare elsewhere means it is a keyword.