A new genetic feature selection with neural network ensemble

A neural network ensemble is a learning paradigm in which a finite collection of neural networks is trained for the same task. Ensembles generally show better classification and generalization performance than a single neural network does. In this paper, a new feature selection method for a neural network ensemble is proposed for pattern classification. The proposed method selects an adequate feature subset for each constituent neural network of the ensemble using a genetic algorithm. Unlike the conventional feature selection method, each neural network is only allowed to have some (not all) of the considered features. The proposed method can therefore be applied to huge-scale feature classification problems. Experiments are performed with four databases to illustrate the performance of the proposed method.     

A parallelized artificial immune network for fuzzy clustering

An artificial immune network (AIN), AINFCM, has been successfully used for fuzzy clustering to overcome the shortage of FCM algorithm that is sensitive to the selection of initial centres. However, as a stochastic searching algorithm, the runtime of AINFCM goes up especially when dealing with large quantities of data or generating much more antibodies for clone selection. In this paper, the PAINFCM is proposed for parallel affinity calculation of antibodies according to time complexity of AINFCM algorithm. Subsequently, a coarse-grained version of PAINFCM algorithm was proposed to parallelize clone expansion. Experiments indicated that the PAINFCM improves efficiency of AIN. Furthermore, it provides a good balance between global searching ability and run time of the AIN.     

Noise analysis compact differential evolution

This article proposes a compact algorithm for optimisation in noisy environments. This algorithm has a compact structure and employs differential evolution search logic. Since it is a compact algorithm, it does not store a population of solutions but a probabilistic representation of the population. This kind of algorithmic structure can be implemented in those real-world problems characterized by memory limitations. The degree of randomization contained in the compact structure allows a robust behaviour in the presence of noise. In addition the proposed algorithm employs the noise analysis survivor selection scheme. This scheme performs an analysis of the noise and automatically performs a re-sampling of the solutions in order to ensure both reliable pairwise comparisons and a minimal cost in terms of fitness evaluations. The noise analysis component can be reliably used in noise environments affected by Gaussian noise which allow an a priori analysis of the noise features. This situation is typical of problems where the fitness is computed by means of measurement devices. An extensive comparative analysis including four different noise levels has been included. Numerical results show that the proposed algorithm displays a very good performance since it regularly succeeds at handling diverse fitness landscapes characterized by diverse noise amplitudes.     

Towards a goal‐driven approach to action selection in self‐adaptive software

Self‐adaptive software is a closed‐loop system, since it continuously monitors its context (i.e. environment) and/or self (i.e. software entities) in order to adapt itself properly to changes. We believe that representing adaptation goals explicitly and tracing them at run‐time are helpful in decision making for adaptation. While goal‐driven models are used in requirements engineering, they have not been utilized systematically yet for run‐time adaptation. To address this research gap, this article focuses on the deciding process in self‐adaptive software, and proposes the Goal‐Action‐Attribute Model (GAAM). An action selection mechanism, based on cooperative decision making, is also proposed that uses GAAM to select the appropriate adaptation action(s). The emphasis is on building a light‐weight and scalable run‐time model which needs less design and tuning effort comparing with a typical rule‐based approach. The GAAM and action selection mechanism are evaluated using a set of experiments on a simulated multi‐tier enterprise application, and two sample ordinal and cardinal action preference lists. The evaluation is accomplished based on a systematic design of experiment and a detailed statistical analysis in order to investigate several research questions. The findings are promising, considering the obtained results, and other impacts of the approach on engineering self‐adaptive software. Although, one case study is not enough to generalize the findings, and the proposed mechanism does not always outperform a typical rule‐based approach, less effort, scalability, and flexibility of GAAM are remarkable. Copyright © 2011 John Wiley & Sons, Ltd.     

Simultaneous estimation and factor selection in quantile regression via adaptive sup-norm regularization

Some regularization methods, including the group lasso and the adaptive group lasso, have been developed for the automatic selection of grouped variables (factors) in conditional mean regression. In many practical situations, such a problem arises naturally when a set of dummy variables is used to represent a categorical factor and/or when a set of basis functions of a continuous variable is included in the predictor set. Complementary to these earlier works, the simultaneous and automatic factor selection is examined in quantile regression. To incorporate the factor information into regularized model fitting, the adaptive sup-norm regularized quantile regression is proposed, which penalizes the empirical check loss function by the sum of factor-wise adaptive sup-norm penalties. It is shown that the proposed method possesses the oracle property. A simulation study demonstrates that the proposed method is a more appropriate tool for factor selection than the adaptive lasso regularized quantile regression.     

Functional linear regression after spline transformation

Functional linear regression has been widely used to model the relationship between a scalar response and functional predictors. If the original data do not satisfy the linear assumption, an intuitive solution is to perform some transformation such that transformed data will be linearly related. The problem of finding such transformations has been rather neglected in the development of functional data analysis tools. In this paper, we consider transformation on the response variable in functional linear regression and propose a nonparametric transformation model in which we use spline functions to construct the transformation function. The functional regression coefficients are then estimated by an innovative procedure called mixed data canonical correlation analysis (MDCCA). MDCCA is analogous to the canonical correlation analysis between two multivariate samples, but is between a multivariate sample and a set of functional data. Here, we apply the MDCCA to the projection of the transformation function on the B-spline space and the functional predictors. We then show that our estimates agree with the regularized functional least squares estimate for the transformation model subject to a scale multiplication. The dimension of the space of spline transformations can be determined by a model selection principle. Typically, a very small number of B-spline knots is needed. Real and simulation data examples are further presented to demonstrate the value of this approach.     

Fast wrapper feature subset selection in high-dimensional datasets by means of filter re-ranking

This paper deals with the problem of supervised wrapper-based feature subset selection in datasets with a very large number of attributes. Recently the literature has contained numerous references to the use of hybrid selection algorithms: based on a filter ranking, they perform an incremental wrapper selection over that ranking. Though working fine, these methods still have their problems: (1) depending on the complexity of the wrapper search method, the number of wrapper evaluations can still be too large; and (2) they rely on a univariate ranking that does not take into account interaction between the variables already included in the selected subset and the remaining ones.Here we propose a new approach whose main goal is to drastically reduce the number of wrapper evaluations while maintaining good performance (e.g. accuracy and size of the obtained subset). To do this we propose an algorithm that iteratively alternates between filter ranking construction and wrapper feature subset selection (FSS). Thus, the FSS only uses the first block of ranked attributes and the ranking method uses the current selected subset in order to build a new ranking where this knowledge is considered. The algorithm terminates when no new attribute is selected in the last call to the FSS algorithm. The main advantage of this approach is that only a few blocks of variables are analyzed, and so the number of wrapper evaluations decreases drastically.The proposed method is tested over eleven high-dimensional datasets (2400-46,000 variables) using different classifiers. The results show an impressive reduction in the number of wrapper evaluations without degrading the quality of the obtained subset.     

基于EM算法的混合模型医学图像分割

医学图像分割中的期望最大化(EM)算法在求解混合模型参数时存在局限性。为此,提出一种模糊约束的混合模型图像分割算法。该算法以像素的独立性假设为前提,在采用EM算法对模型参数进行求解的过程中,通过模糊集合论方法,引入像素空间信息。实验结果表明,该算法没有引入新的模型参数,能够保持独立混合模型的简单性,且具有自动模型选择能力,可以获得较理想的分割结果。     

某深基坑工程方案设计比选分析

某深基坑工程地下水资源丰富,周边环境非常复杂.针对本工程的现场实际情况,介绍了复杂环境条件下深基坑支护方案设计比选的要点,并结合当地的施工经验来选择支护方案,给类似项目提供了一种方案比选方法,具有一定的实用价值.     

钢管厂双层异形网壳结构的设计与分析研究

以某钢管厂双层异形网壳结构设计为工程背景,对其结构选型做了具体论述;以线弹性分析作为基础,细致地分析了异形双层柱面网壳结构的非线性性能.最后得出网壳结构工程设计与研究的几点结论.