Learn R Programming

clustComp (version 1.0.0)

Clustering Comparison Package

Description

clustComp is a package that implements several techniques for the comparison and visualisation of relationships between different clustering results, either flat versus flat or hierarchical versus flat. These relationships among clusters are displayed using a weighted bi-graph, in which the nodes represent the clusters and the edges connect pairs of nodes with non-empty intersection; the weight of each edge is the number of elements in that intersection and is displayed through the edge thickness. The best layout of the bi-graph is provided by the barycentre algorithm, which minimises the weighted number of crossings. In the case of comparing a hierarchical and a non-hierarchical clustering, the dendrogram is pruned at different heights, selected by exploring the tree by depth-first search, starting at the root. Branches are decided to be split according to the value of a scoring function, that can be based either on the aesthetics of the bi-graph or on the mutual information between the hierarchical and the flat clusterings. A mapping between groups of clusters from each side is constructed with a greedy algorithm, and can be additionally visualised.

Copy Link

Version

Version

1.0.0

License

GPL (>= 2)

Maintainer

Aurora Torrente

Last Published

February 15th, 2017

Functions in clustComp (1.0.0)

flatVSflat

Comparison of two flat clusterings
score.it

Computation of the information theoretic-based score of the parent and the children trees
dyn.cross

Computation of the number of crossings in the bi-graph in a particular layout
drawTreeGraph

Plot the bi-graph determined by the branches in the tree and the flat clusters
score.crossing

Computation of the aesthetics-based score of the parent and the children trees
barycentre

Computation of the barycentre-coordinate of a node connected to others in a bigraph
flatVShier

Comparison of a hierarchical and a flat clusterings
insert

Insert a set of values at a given position of a vector
SCmapping

Construction of the superclusters and the one-to-one mapping between them