AI engineering glossary · NLP embeddings
What is distributional hypothesis?
Distributional hypothesis: The idea, often summarised as "a word is known by the company it keeps", that words appearing in similar contexts have similar meanings. It is the principle behind almost every method that learns word vectors from raw text.
Distributional hypothesis: an example
"doctor" and "physician" both appear near words like hospital, patient and prescribe, so they get similar vectors.
Related terms
More NLP embeddings terms
- Document-term matrix
- Dot product similarity
- ELMo
- Entity linking
- Entity-level F1
- FastText
- Fuzzy matching
- Gazetteer
Where you learn it
Distributional hypothesis is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.