Title :
An online clustering algorithm
Author :
Kan Li ; Fenglan Yao ; Ruipeng Liu
Author_Institution :
Sch. of Comput., Beijing Inst. of Technol., Beijing, China
Abstract :
This paper presents a new online clustering algorithm called SAFN which is used to learn continuously evolving clusters from non-stationary data. The SAFN uses a fast adaptive learning procedure to take into account variations over time. In non-stationary and multi-class environment, the SAFN learning procedure consists of five main stages: creation, adaptation, mergence, split and elimination. Experiments are carried out in three kinds of datasets to illustrate the performance of the SAFN algorithm for online clustering. Compared with SAKM algorithm, SAFN algorithm shows better performance in accuracy of clustering and multi-class high-dimension data.
Keywords :
adaptive systems; learning (artificial intelligence); pattern clustering; SAFN learning procedure; SAKM algorithm; fast adaptive learning procedure; multi class environment; multi class high dimension data; nonstationary data; online clustering algorithm; Clustering algorithms; Gaussian distribution; Kernel; Neurons; Noise; Signal processing algorithms; Support vector machines; Online clustering; non-stationary data; self-adaptive feed-forward neural network; similarity measure;
Conference_Titel :
Fuzzy Systems and Knowledge Discovery (FSKD), 2011 Eighth International Conference on
Conference_Location :
Shanghai
Print_ISBN :
978-1-61284-180-9
DOI :
10.1109/FSKD.2011.6019762
