Nearest neighbor clustering algorithm based on AIC criterion;

First, method for initialization of wavelet basis in wavelet frames is proposed; second, according to the data distribution, a modified method for rough selection of wavelet basis is given; and the last, an adaptive projection (algorithm) combined with AIC criterion is used to purify the wavelet basis, meanwhile finishing the parameters (identification.

This paper uses the AIC criterion proposed by Akaike(1973) to choose the common factor s number in factor analysis model,answers the question at the end of paper[1] and gives some discussion results, also analyzes and studies the data in a comparative study in measuring stress among university faculty in China and Japan.

In this paper, wind speed time series simulation methods and AIC rule is elaborate.

The ship longitudinal motion was simulated in the typical voyages condition, and it was compared with AIC rule order selection algorithm.

In this paper, based on the natural wind properties and the correlativity of nodal wind speed time series, AR model is employed to simulate wind speed time series, and the AIC rule is used to fix the rank of the AR model.

The model structure is determined by AIC and stepbystep regression arithmetic, parameters of the model are estimated by the least square method.

A method of confirming the coefficient of confidence is presented with AIC criteria.

From the basic MDL criteria, and considering the advantages and shortfalls of the AIC criteria and MDL criteria, a new improved algorithm is presented in this paper.