The cosine similarity measures how close two vectors are to one another using the cosine of the angle between them. Unlike a Euclidean distance, the cosine similarity ignores the size of the two vectors, it just focuses on their direction. In {{document retrieval}] and text processing, the cosine silarity is applied to the vector representing the frequencies of words in a document, soemtimes weighted by overalll corpus frequency.
Defined on page 213
Used on page 213