Method for improving classification performance of neural network based on fuzzy input and network inversion

In order to effectively improve the classification performance of neural network, first architecture of fuzzy neural network with fuzzy input was proposed. Next a cost function of fuzzy outputs and non-fuzzy targets was defined. Then a learning algorithm from the cost function for adjusting weights was derived. And then the fuzzy neural network was inversed and fuzzified inversion algorithm was proposed. Finally, computer simulations on real-world pattern classification problems examine the effectives of the proposed approach. The experiment results show that the proposed approach has the merits of high learning efficiency, high classification accuracy and high generalization capability.

Construction of coordination networks with high connectivity: a new 8-connected self-penetrating network based on tetranuclear metal clusters

Two highly connected cobalt(II) and zinc(II) coordination polymers with tetranuclear metal clusters as the nodes of network have been prepared, being the first example of an 8-connected self-penetrating net based on a cross-linked alpha-Po subnet.

A neural network based tool for semi-automatic code transformation

A neural network based tool has been developed to assist in the process of code transformation. The tool offers advice on appropriate transformations within a knowledge-driven, semi-automatic parallelisation environment. We have identified the essential characteristics of codes relevant to loop transformations. A Kohonen network is used to discover structure in the characterised codes thus revealing new knowledge that may be brought to bear on the mapping between codes and transformations or transformation sequences. A transform selector based on this process has been developed and successfully applied to the parallelisation of sequential codes.

Neural network based control for synchronous generators

In this paper, a Radial Basis Function neural network based AVR is proposed. A control strategy which generates local linear models from a global neural model on-line is used to derive controller feedback gains based on the Generalised Minimum Variance technique. Testing is carried out on a micromachine system which enables evaluation of practical implementation of the scheme. Constraints imposed by gathering training data, computational load, and memory requirements for the training algorithm are addressed.

Local network-based measures to assess the inferability of different regulatory networks

The purpose of this study is to compare the inferability of various synthetic as well as real biological regulatory networks. In order to assess differences we apply local network-based measures. That means, instead of applying global measures, we investigate and assess an inference algorithm locally, on the level of individual edges and subnetworks. We demonstrate the behaviour of our local network-based measures with respect to different regulatory networks by conducting large-scale simulations. As inference algorithm we use exemplarily ARACNE. The results from our exploratory analysis allow us not only to gain new insights into the strength and weakness of an inference algorithm with respect to characteristics of different regulatory networks, but also to obtain information that could be used to design novel problem-specific statistical estimators.

Identifying Critical Financial Networks of the DJIA: Toward a Network-Based Index

The purpose of this article is twofold. First, we introduce a novel definition of financial networks obtained from time series data from the stock market. Second, we demonstrate that these networks can be used as an index with the property to reflect critical states of the market, respectively, crashes sufficiently. Our work aims to advocate a network-based analysis in the context of the stock market, because such a collective phenomenon can not only be economically described by networks but also analyzed as demonstrated in this article. (C) 2010 Wiley Periodicals, Inc. Complexity 16: 24-33, 2010

A cleavable network based on crosslinked star polymers containing acid-labile diacetal crosslinks: Synthesis, characterization and hydrolysis

A hydrolyzable model network comprising interconnected star polymers was prepared by the sequential group transfer polymerization of methyl methacrylate and the acid-labile diacetal-based dimethacrylate crosslinker bis[(2-methacryloyloxy)ethoxymethyl] ether. in contrast to other polymer networks previously synthesized by our group, all the branching points of this polymer network were found to hydrolyze under mildly acidic conditions, giving a linear copolymer with the theoretically expected molecular weight and composition. The ease of hydrolysis of this polymer network renders it a good candidate for use in the biomedical field. The characterization of the synthesized network, its linear and star polymer precursors and the hydrolysis products of the network and its precursors, by a variety of techniques, established the successful synthesis and hydrolysis of this well-defined polymer nanostructure.