Intermediate · NLP

Vector Representation

Visual diagram · (in preparation) · Math · (in preparation) · Worked example · 3 difficulty levels.

TL;DR. Encoding objects (words, images, users) as numerical vectors so machines can compute similarity and patterns.

Technical Definition

Encoding objects (words, images, users) as numerical vectors so machines can compute similarity and patterns.

How it works

Vector representations underpin embeddings, retrieval, recommendation, and clustering. Similar objects produce nearby vectors under some distance metric (cosine, dot product, Euclidean). The dimensionality and the training objective determine what 'similar' means.

Related Concepts

• Embedding — A dense vector representation that captures semantic meaning, mapping discrete items like words into continuous mathematical space.
• Vector Database — A specialized database optimized for storing, indexing, and querying high-dimensional embedding vectors using similarity search.
• Tokenization — The process of breaking text into smaller units (tokens) that language models can process as numerical inputs.