Sparse Retrieval
A traditional search methodology that relies on exact keyword matching, representing documents as high-dimensional but mostly empty vectors.
Think of It Like This
Like looking up a specific term in a textbook's index to find exactly which pages contain that exact word.
Algorithms like BM25 fall under sparse retrieval. They map documents into a vocabulary space of millions of words, where almost all values are zero except for the words actually present. While they struggle with synonyms and semantic meaning, they are unmatched for precise queries like finding a specific part number or a person's exact name.