TF-IDF
Technology & Computing abbreviation · Artificial Intelligence
| Stands for | Term frequency-inverse document frequency |
|---|---|
| Categories | Technology & Computing |
TF-IDF stands for Term frequency-inverse document frequency, an artificial intelligence abbreviation.
A statistic that scores how important a word is to a document within a collection: words frequent in one document but rare across the collection score highest. It is a classic technique for search ranking, keyword extraction and text classification.
Example: TF-IDF ranks "refund" highly in complaint emails because it rarely appears in other messages.
Usage notes
Karen Spärck Jones introduced inverse document frequency in 1972. The related BM25 formula still powers keyword search in Elasticsearch and Lucene and is often combined with embeddings in hybrid search. Very common words such as "the" receive near-zero weight automatically.
Related abbreviations
More artificial intelligence abbreviations
- SLM – Small language model
- SOTA – State of the art
- STT – Speech-to-text
- SVM – Support vector machine
- t-SNE – t-distributed stochastic neighbor embedding
- TF – TensorFlow
- TTS – Text-to-speech
- VAE – Variational autoencoder
- ViT – Vision Transformer
- VLM – Vision-language model
- XAI – Explainable artificial intelligence
- YOLO – You Only Look Once