About distributional semantics

The basic idea of distributional semantics can be summarized in the so-called distributional hypothesis: "linguistic elements with similar distributions have similar meanings". This means that words with similar distributions are close to each other semantically. In this way, semantic similarity can be easily calculated.

For example, if we discover that two words W1 and W2 tend to have similar distributional properties, such as that both generally appear next to another W3 entity in different texts, and then we could postulate that W1 and W2 belong to the same linguistic class and have a meaning similar. The underlying idea that "a word is defined by the words that the accompany" was popularized by J.R. Firth