Paper Type |
Opinion |
Title |
Parallel K-means Clustering Algorithm |
Author |
Sanpawat Kantabutra |
Email |
- |
Abstract: Despite its simplicity and its linear time, a serial K-means algorithm’s time complexity remains expensive when it is applied to a problem of large size of multidimensional vectors. In this paper the author proposes an improvement by a factor of O(K/2) where K is the number of desired clusters, by applying theories of parallel computing to the algorithm. In addition to time improvement, the parallel version of K-means algorithm also enables the algorithm to run on larger memory of multiple machines when the memory of a single machine is insufficient to solve a problem.
|
|
Start & End Page |
- |
Received Date |
1999-04-02 |
Revised Date |
|
Accepted Date |
1999-08-05 |
Full Text |
Download |
Keyword |
|
Volume |
Vol.26 No.2 (DECEMBER 1999) |
DOI |
|
Citation |
Kantabutra S., Parallel K-means Clustering Algorithm, Chiang Mai J. Sci., 1999; 26(2): -. |
SDGs |
|
View:602 Download:77 |