Tag: search-ranking

Learning to rank reframed search ranking as supervised machine learning, allowing systems to combine many relevance signals using models trained on preference judgments rather than fixed hand-tuned formulas.

BM25 grew from Stephen Robertson's probabilistic retrieval research and the Okapi system, combining term rarity, saturating term frequency and document-length normalization.

Gerard Salton's SMART research and the vector space model turned documents and queries into weighted term vectors, making graded similarity and ranked retrieval central to search.