Paper
Influence on ART2 Clustering Algorithm with Different Adjusting Learning Rate
- Authors:
- Shujie Du
- Abstract
- The process of ART2 can recognize learned models fast and be adapted to new objects rapidly. It carries out clustering with hierarchy structure by using competitive learning and self-steady mechanism in dynamic environment with noise and without supervision. Here discuss the common-used learning rules at first. The way to adjust learning rate is suggested and the assimilation effect is verified by a shape learning trial. The categorization results are also compared to illustrate the effects of different learning rates. To some extent, the improved algorithm solves the pattern drifting problem.
- Keywords
- ART2; Assimilation Effect; Data Clustering; Learning Rate; Adaptation Process
- StartPage
- 30
- EndPage
- 34
- Doi