Karen Spärck Jones
Inverse Document Frequency
The weighting that made search work
Rating
90Impact score
Domains
Information retrievalNatural language processingSearch
Scouting report
Her 1972 paper introduced inverse document frequency: rare terms tell you more than common ones. TF-IDF underpinned search engines for thirty years and is still the baseline every ranking system is measured against. She argued throughout her career that computing was too important to be left to men, and said so in those words.
Curator’s note
One weighting function, and search stopped being useless.
Portrait
Painted from a photograph of Karen Spärck Jones by en:user:Markus Kuhn, licensed CC BY 2.5.
Sources
- 01Karen Spärck Jones en.wikipedia.org
- 02A statistical interpretation of term specificity (1972) emerald.com