Manning, Christopher, Prabhakar Raghavan and Hinrich Schütze. 2008. “Flat Clustering” and “Hierarchical Clustering.” Chapters 16 and 17 from Introduction to Information Retrieval.
The intuition behind clustering is that we want a collection of documents to fall into separate, internally homogeneous clusters. This seems like a reasonable desire if you want to see how a collection of documents is distinct, but what about when we want to see how/where a collection of documents overlaps? If clustering won't help us here, than what other methods are available to us?
The intuition behind clustering is that we want a collection of documents to fall into separate, internally homogeneous clusters. This seems like a reasonable desire if you want to see how a collection of documents is distinct, but what about when we want to see how/where a collection of documents overlaps? If clustering won't help us here, than what other methods are available to us?