Canopy clustering algorithm

It is intended to speed up clustering operations on large data sets, where using another algorithm directly may be impractical because of the size of the data set.

The algorithm proceeds as follows:

• Cheaply partition the data into overlapping subsets, called 'canopies'
• Perform more expensive clustering, but only within these canopies

Benefits

• The number of instances of training data that must be compared at each step is reduced
• There is some evidence that the resulting clusters are improved

References

McCallum, Nigamy and Ungar: "Efficient Clustering of High Dimensional Data Sets with Application to Reference Matching"

External links

• Cluster Computing and MapReduce Lecture 4 from Google

See also

• Data clustering
• K-means algorithm
• Linde-Buzo-Gray algorithm