|
Published Articles >> Table of Contents >> Abstract
15th International Conference on Pattern Recognition (ICPR'00) - Volume 2
p. 2076
Clustering Very Large Databases Using EM Mixture Models
P.S. Bradley, Microsoft Research
C.A. Reina, Microsoft Research
U.M. Fayyad, digiMine.com
Full Article Text:
  
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/ICPR.2000.906021
Send link to a friend
| Abstract |
|
Clustering very large databases is a challenge for traditional pattern recognition algorithms, e.g. the Expectation-Maximization (EM) algorithm for fitting mixture models, because of high memory and iteration requirements. Over large databases, the cost of the numerous scans required converging and large memory requirement of the algorithm becomes prohibitive. We present a decomposition of the EM algorithm requiring a small amount of memory by limiting iterations to small data subsets. The scalable EM approach requires at most one database scan and is based on identifying regions of the data that are discardable, regions that are compressible, and regions that must be maintained in memory. Data resolution is preserved to the extent possible based upon the size of the memory buffer and fit of the current model to the data. Computational tests demonstrate that the scalable scheme outperforms similarly constrained EM approaches.
|
 Additional Information
|
Citation:
P.S. Bradley, C.A. Reina, U.M. Fayyad,
"Clustering Very Large Databases Using EM Mixture Models,"
icpr,
p. 2076,
15th International Conference on Pattern Recognition (ICPR'00) - Volume 2,
2000
|
|
