Nearest Neighbors

Find nearest neighbors using exhaustive search or Kd-tree search

A Nearest neighbor search locates the k-nearest neighbors or all neighbors within a specified distance to query data points, based on the specified distance metric. Available distance metrics include Euclidean, Hamming, and Mahalanobis, among others.

Statistics and Machine Learning Toolbox™ offers two ways to find nearest neighbors. You can create a searcher object with a training data set, and pass the object and query data sets to the object functions (knnsearch and rangesearch). Or, you can use the knnsearch and rangesearch functions, which take both a training data set and a query data set directly. Creating a searcher object is preferable when you have multiple query data sets because the searcher object stores information common to the data sets. For example, a KDTreeSearcher object stores a Kd-tree.

Blocks

KNN Search

Find k-nearest neighbors using searcher object (Since R2023b)

Functions

expand all

Create Nearest Neighbor Searcher Object

`ExhaustiveSearcher`	Create exhaustive nearest neighbor searcher
`KDTreeSearcher`	Create Kd-tree nearest neighbor searcher
`hnswSearcher`	Hierarchical Navigable Small Worlds (HNSW) approximate nearest neighbor search (Since R2024a)
`createns`	Create nearest neighbor searcher object

Find Nearest Neighbors

Find Neighbors Using Searcher Object

`knnsearch`	Find k-nearest neighbors using searcher object
`rangesearch`	Find all neighbors within specified distance using searcher object

Find Neighbors Using Input Data

`knnsearch`	Find k-nearest neighbors using input data
`rangesearch`	Find all neighbors within specified distance using input data

Topics

Classification Using Nearest Neighbors
Categorize data points based on their distance to points in a training data set, using a variety of distance metrics.