# Package elki.clustering.kmedoids

K-medoids clustering (PAM).
• Interface Summary
Interface Description
KMedoidsClustering<O>
Interface for clustering algorithms that produce medoids.
• Class Summary
Class Description
AlternatingKMedoids<O>
A k-medoids clustering algorithm, implemented as EM-style batch algorithm; known in literature as the "alternate" method.
AlternatingKMedoids.Par<V>
Parameterization class.
CLARA<V>
Clustering Large Applications (CLARA) is a clustering method for large data sets based on PAM, partitioning around medoids (PAM) based on sampling.
CLARA.CachedDistanceQuery<V>
Cached distance query.
CLARA.Par<V>
Parameterization class.
CLARANS<O>
CLARANS: a method for clustering objects for spatial data mining is inspired by PAM (partitioning around medoids, PAM) and CLARA and also based on sampling.
CLARANS.Assignment
Assignment state.
CLARANS.Par<V>
Parameterization class.
EagerPAM<O>
Variation of PAM that eagerly performs all swaps that yield an improvement during an iteration.
EagerPAM.Instance
Instance for a single dataset.
EagerPAM.Par<O>
Parameterization class.
FastCLARA<V>
Clustering Large Applications (CLARA) with the FastPAM improvements, to increase scalability in the number of clusters.
FastCLARA.Par<V>
Parameterization class.
FastCLARANS<V>
A faster variation of CLARANS, that can explore O(k) as many swaps at a similar cost by considering all medoids for each candidate non-medoid.
FastCLARANS.Assignment
Assignment state.
FastCLARANS.Par<V>
Parameterization class.
FasterCLARA<O>
Clustering Large Applications (CLARA) with the FastPAM improvements, to increase scalability in the number of clusters.
FasterCLARA.Par<V>
Parameterization class.
FasterPAM<O>
Variation of FastPAM that eagerly performs any swap that yields an improvement during an iteration.
FasterPAM.Instance
Instance for a single dataset.
FasterPAM.Par<O>
Parameterization class.
FastPAM<O>
FastPAM: An improved version of PAM, that is usually O(k) times faster.
FastPAM.Instance
Instance for a single dataset.
FastPAM.Par<V>
Parameterization class.
FastPAM1<O>
FastPAM1: A version of PAM that is O(k) times faster, i.e., now in O((n-k)²).
FastPAM1.Instance
Instance for a single dataset.
FastPAM1.Par<V>
Parameterization class.
PAM<O>
The original Partitioning Around Medoids (PAM) algorithm or k-medoids clustering, as proposed by Kaufman and Rousseeuw; a largely equivalent method was also proposed by Whitaker in the operations research domain, and is well known by the name "fast interchange" there.
PAM.Instance
Instance for a single dataset.
PAM.Par<O>
Parameterization class.
ReynoldsPAM<O>
The Partitioning Around Medoids (PAM) algorithm with some additional optimizations proposed by Reynolds et al.
ReynoldsPAM.Instance
Instance for a single dataset.
ReynoldsPAM.Par<V>
Parameterization class.
SingleAssignmentKMedoids<O>
K-medoids clustering by using the initialization only, then assigning each object to the nearest neighbor.
SingleAssignmentKMedoids.Instance
Instance for a single dataset.
SingleAssignmentKMedoids.Par<O>
Parameterization class.