execute_package_parallel {Clustering} | R Documentation |
Evaluation clustering algorithm.
Description
Method that evaluates clustering algorithm from a file directory or dataframe.
Usage
execute_package_parallel(
directory_files,
df,
algorithms_execute,
measures_execute,
cluster_min,
cluster_max,
metrics_execute,
attributes,
number_algorithms,
numberClusters,
numberDataSets,
is_metric_external,
is_metric_internal,
name_dataframe
)
Arguments
directory_files |
It's a string with the route where the datasets are located. |
df |
Data matrix or data frame, or dissimilarity matrix, depending on the value of the argument. |
algorithms_execute |
Character vector with the algorithms to be executed. The algorithms implemented are: hclust, apclusterK,agnes,clara,daisy,diana,fanny,mona,pam,gmm,kmeans_arma, kmeans_rcpp,mini_kmeans, pvclust. |
measures_execute |
Character array with the measurements of dissimilarity to be executed. Depending on the algorithm, one or the other is implemented. Among them we highlight: Euclidena, Manhattan, etc. |
cluster_min |
Minimum number of clusters. |
cluster_max |
Maximum number of clusters. cluster_max must be greater or equal cluster_min. |
metrics_execute |
Character array defining the metrics to be executed. The night metrics implemented are: Entropy, Variation_information, Precision, Recall, F_measure, Fowlkes_mallows_index, Connectivity, Dunn and Silhouette. |
number_algorithms |
It's a numeric field with the number of algorithms. |
numberClusters |
It's a numeric field with the difference between clusters. |
numberDataSets |
It's a numeric field with the number of datasets. |
is_metric_external |
Boolean field to indicate whether to run external metrics. |
is_metric_internal |
Boolean field to indicate whether to run internal metrics. |
name_dataframe |
Name of data.frame when is fill. |
Value
Returns a list with the result matrix of evaluating the data from the indicated algorithms, metrics and number of clusters.