Alternative Clustering
From ResearchWiki
Clustering is a challenge whose difficulty is caused by the lack of a unique and precise definition of what a cluster is. Information is not available about the underlying structure of data, nor is there a unique similarity measure for differentiating clusters. Hence, there is often no single clustering solution that explains the structure of a given dataset, especially if it is complex and represented in high dimensional space.
- This challenge has given rise to the recently emerging area of alternative clustering analysis, whose goal is to seek different partitions( or clusterings), in order to describe different grouping aspects for a given dataset.
Recently a novel algorithm called CAMI [1] has been proposed, which can uncover alternative clusterings from a dataset why is the page not getting saved??????<ref>i hate this</ref> m hating this
