kmknn: k-Means for k-Nearest Neighbors

Implements a k-nearest neighbor search based on pre-clustering with the k-means algorithm, as described by Wang (2012). This is faster than conventional kd-trees for neighbor searching in higher (> 20) dimensional data. Functions are also provided to search for all neighbors within a given distance. Parallelization is achieved using the BiocParallel framework.

Package details

AuthorAaron Lun [aut, cre, cph]
Bioconductor views Classification Clustering
MaintainerAaron Lun <[email protected]>
Package repositoryView on Bioconductor
Installation Install the latest version of this package by entering the following in R:

Try the kmknn package in your browser

Any scripts or data that you put into this service are public.

kmknn documentation built on Nov. 1, 2018, 4:21 a.m.