| 摘要 |
PROBLEM TO BE SOLVED: To improve cluster quality when data substantially proceeds with a lapse of time. SOLUTION: In regard to a technique performing data clustering of a data stream, first, online statistics are generated from a data stream. Thereafter, offline processing of the online statistics is performed when the offline processing is needed or desired. The online statistics can be generated through the reception of data points from the data stream, and the formation and the update of a data group. The offline processing can be performed by re-clustering the data point group in the periphery of the sampled data points, and a newly formed cluster is reported. COPYRIGHT: (C)2005,JPO&NCIPI
|
