You should be able to answer each of the following questions.
- How are clustering of documents and clustering of terms
related?
- What good is a hierarchical cluster tree to a IS &R system?
- What effect will choice of similarity function have on the
end result of clustering?
- Is it preferable to use a partitioning method or an
agglomerative method? Why?
- What is the right clustering algorithm to use, given a
particular collection of data?