Improved Fuzzy Clustering Algorithm Based on Intelligent Computing

FCM algorithm is easy to be affected by fuzzy parameters, initial clustering, noise, and because of the single iteration path, the phenomenon of local extremum can be generated. Although it was improvd by artificial bee colony algorithm, artificial bee colony has some shortcomings, such as the prema...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	2017 International Conference on Robots & Intelligent System (ICRIS) S. 161 - 164
1. Verfasser:	Zhu Tianyuan
Format:	Tagungsbericht
Sprache:	Englisch
Veröffentlicht:	IEEE 01.10.2017
Schlagworte:	artificial bee colony clustering center fuzzy c-means Intelligent systems Robots Xenon
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Beschreibung
Zusammenfassung:	FCM algorithm is easy to be affected by fuzzy parameters, initial clustering, noise, and because of the single iteration path, the phenomenon of local extremum can be generated. Although it was improvd by artificial bee colony algorithm, artificial bee colony has some shortcomings, such as the premature convergence, low precision and slow convergence. A kind of improved artificial swarm algorithm based on improved search strategy is proposed and is also used in fuzzy c-means algorithm. The experiments based on UCI datasets show that this new algorithm overcomes the disadvanhges of FCM. Besides, it has higher clustering accuracy rate.
DOI:	10.1109/ICRIS.2017.47