Vector search is a technique used in natural language processing (NLP) to find similar data points in high-dimensional vector spaces. It uses mathematical representations called vectors to store and retrieve information. Vectors are arrays of floating-point numbers that capture the essence of input data, such as text or images. Vector search retrieves relevant data that answers an input query by finding vectors with small distances from a query vector. Dense vectors are used for semantic search, while sparse vectors are used for lexical search. Popular algorithms for performing vector search include brute-force, Inverted File Index (IVF), Hierarchical Navigable Small Worlds (HNSW), and quantization. Vector search is essential for neural information retrieval systems, and various libraries, databases, and frameworks have been developed to support it. The technique has many applications, including search engines, recommendation systems, and content-based recommender systems.