word2vec
Definition
word2vec learns numerical vectors for words from their linguistic context.
A model and tool collection for learning dense word vectors with CBOW or Skip-gram from local contexts.
Simply explained
Words get numerical addresses. Words that occur in similar contexts can thus receive similar addresses.
Why it matters in practice
The term helps to select, develop, evaluate AI systems in a well-founded manner and realistically assess their technical risks.
Example from business
A company analyzes technical term similarities but checks for bias and rare words.
Limits and pitfalls
- word2vec provides context-independent word vectors; tokenization and corpus shape meaning and biases.
Origin & history
Mikolov and co-authors published word2vec in 2013 and made efficient word representations widely available.
Use in mid-sized companies
- Possible example: A company analyzes technical term similarities but checks for bias and rare words.
Frequently asked questions
What does word2vec mean?
A model and tool collection for learning dense word vectors with CBOW or Skip-gram from local contexts. What should you watch out for? word2vec provides context-independent word vectors; tokenization and corpus shape meaning and biases.
Sources
- Efficient Estimation of Word Representations · Mikolov et al., Efficient Estimation of Word Representations
Source checked on 10.10.2026. Editorial application examples are hypothetical; no evidence of real customer results.
Editorially reviewed on: