Package
weka.clusterers
Synopsis
Hierarchical clustering class.
Implements a number of classic agglomorative (i.e. bottom up) hierarchical clustering methods.
Options
The table below describes the options available for HierarchicalClusterer.
Option | Description |
---|---|
debug | If set to true, classifier may output additional info to the console. |
distanceFunction | Sets the distance function, which measures the distance between two individual. instances (or possibly the distance between an instance and the centroid of a cluster depending on the Link type). |
linkType | Sets the method used to measure the distance between two clusters. |
numClusters | Sets the number of clusters. If a single hierarchy is desired, set this to 1. |
printNewick | Flag to indicate whether the cluster should be print in Newick format. This can be useful for display in other programs. However, for large datasets a lot of text may be produced, which may not be a nuisance when the Newick format is not required |
Capabilities
The table below describes the capabilities of HierarchicalClusterer.
Capability | Supported |
---|---|
Class | No class |
Attributes | String attributes, Date attributes, Missing values, Unary attributes, Numeric attributes, Binary attributes, Nominal attributes, Empty nominal attributes |
Min # of instances | 0 |