DeltaDens - incremental algorithm for on-line density-based clustering
[ 1 ] Instytut Informatyki, Wydział Informatyki, Politechnika Poznańska | [ P ] employee
2013
paper
english
- Density–based Clustering
- On–line Clustering
- Data Streams
EN Cluster analysis of data delivered in a stream exhibits some unique properties. They make the clustering more difficult than it happens for the static set of data. This paper introduces a new DeltaDens clustering algorithm that can be used for this purpose. It is a density–based algorithm, capable of finding an unbound number of irregular clusters. The algorithm’s per–iteration processing time linearly depends on the size of its internal buffer. The paper describes the algorithm and delivers some experimental results explaining its performance and accuracy.
163 - 172
WoS (15)