Hierarchical Clustering Analysis

Description
  • These are all non-hypothesis techniques.
  • Clustering is achieved on the basis of a defined or measured 'distance' or 'similarity' between individuals / objects.
  • Those objects with a high degree of 'similarity' will be placed in the same 'cluster'.
  • Those items that appear 'distant' or dissimilar from the first group will be placed in another cluster.
  • Eventually all individuals / objects will be allocated to a cluster. The effect is to produce a hierarchy of clusters and these can be represented by a dendrogram.




Usage

Field Description
Launch Field indicating whether you want to execute this analysis in the workflow (checked) or not (unchecked).
Distance Measure of dissimilarity between sets of observations
Agglomeration Method Function of the pairwise distances between observations
Reverse HeatMap By default, for the representation of the HeatMap, the samples are in the height direction, the variables in the direction of the width. You can reverse the Heat Map by checking this option at Yes




Output Examples

default/hca.txt ยท Last modified: 2014/12/03 15:11 by admin