tNN-class: Class "tNN"

In rEMM: Extensible Markov Model for Modelling Temporal Relationships Between Clusters

Class "tNN"

Description

Implements the threshold Nearest Neighbor clustering algorithm used by EMM.

Objects from the Class

Objects can be created with new() or by the creator function tNN.

Slots

measure:

Object of class "character" containing the name of the dissimilarity measure used (see dist in proxy for available measures)

centroids:

Object of class "logical" indicating if centroids are used for clusters. If FALSE, pseudo medians (first observation of a cluster) are used to represent a cluster.

threshold:

Object of class "numeric" with the dissimilarity threshold used by the NN clustering algorithm for assigning a new observation to existing clusters.

lambda:

Object of class "numeric" specifying the rate for fading.

lambda_factor:

Object of class "numeric" expressing the fading rate expressed as a factor.

tnn_d:

An environment containing the variable data for the tNN object:

centers:

Object of class "matrix" containing the cluster centers.

counts:

Object of class "numeric" with the number of observations assigned to each cluster.

var_thresholds:

Object of class "numeric" with the dissimilarity thresholds for individual clusters (usually the same as threshold).

last:

A "character" vector containing the cluster names the points for the previous call of cluster() were assigned to.

Methods

copy

signature(x = "tNN"): Make a copy of the tNN object. Making explicit copies is necessary since information is stored in an environment which is not copied for regular assignements.

cluster_centers

signature(x = "tNN"): returns the cluster centers as a matrix.

cluster_counts

signature(x = "tNN"): returns the cluster counts as a vector.

clusters

signature(x = "tNN"): returns the names of the clusters.

last_clustering

signature(x = "tNN"): returns the indices of the clusters the data points in the last cluster operation where assigned to. To save memory the last clustering information can be removed by setting the formal parameter remove to TRUE.

nclusters

signature(x = "tNN"): returns the number of clusters in the clustering.

plot

signature(x = "tNN", y = "missing"): plots the cluster centers using a scatterplot matrix (see pairs).

References

M.H. Dunham, Y. Meng, J. Huang (2004): Extensible Markov Model, In: ICDM '04: Proceedings of the Fourth IEEE International Conference on Data Mining, pp. 371–374.

See Also

cluster for adding new data to the clustering. find_clusters to find the nearest neighbor cluster for given data points. EMM extends "tNN".

rEMM documentation built on May 29, 2024, 4:35 a.m.