A New Fuzzy Entropy Clustering Method with Controllable Membership Characteristics

Yang, Dian-Rong; Lan, Leu-Shing; Pao, Wei-Cheng

doi:10.1109/mwscas.2006.382028

Cited by 5 publications

(1 citation statement)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this paper, the entropy clustering theory is introduced to solve these problems. The advantage of entropy clustering is that the number of basis function can be obtained by analyzing the input data and classifying the input data according to entropy at each data point [12], [13], [14]. If choosing radial basic function as function cluster of functional network, we can get not only the number of basis function but also initial value of the center [7], [8], [9].…”

Section: Introductionmentioning

confidence: 99%

Optimizing Neuron Function Based on Entropy Clustering in Functional Networks

Yujiong

Zhou

2008

2008 International Conference on Computational Intelligence and Security

View full text Add to dashboard Cite

Functional networks are the extension of neural networks which have been studied recently. Like neural networks, there is no systematic method for designing approximation functional network structures. In this paper, a new entropy clustering method designed for functional networks is presented, which combines each neuron function and functional parameters by performing the optimal search to achieve the learning between functional network structures and the functional parameters. The simulation results indicate that the proposed method can produce more rational structure and greatly improve convergent precision of functional networks.

show abstract