A new ensemble feature selection and its application to pattern classification

Neural network ensemble based on rough sets reduct is proposed to decrease the computational complexity of conventional ensemble feature selection algorithm. First, a dynamic reduction technology combining genetic algorithm with resampling method is adopted to obtain reducts with good generalization ability. Second, Multiple BP neural networks based on different reducts are built as base classifiers. According to the idea of selective ensemble, the neural network ensemble with best generalization ability can be found by search strategies. Finally, classification based on neural network ensemble is implemented by combining the predictions of component networks with voting. The method has been verified in the experiment of remote sensing image and five UCI datasets classification. Compared with conventional ensemble feature selection algorithms, it costs less time and lower computing complexity, and the classification accuracy is satisfactory.     

Speed up Training of the Recurrent Neural Network Based on Constrained optimization Techniques

In this paper,the constrained optimization technique for a substantial problem is explored,that is accelerating training the globally recurrent neural network.Unlike most of the previous methods in feedforware neural networks,the authors adopt the constrained optimization technique to improve the gradientbased algorithm of the globally recurrent neural network for the adaptive learning rate during tracining.Using the recurrent network with the improved algorithm,some experiments in two real-world problems,namely,filtering additive noises in acoustic data and classification of temporat signals for speaker identification,have been performed.The experimental results show that the recurrent neural network with the improved learning algorithm yields significantly faster training and achieves the satisfactory performance.     

Nonlinear optimized adaptive trajectory control of helicopter

This paper attempts to develop an optimized adaptive trajectory control system for helicopters based on the dynamic inversion method. This control algorithm is implemented by three time-scale separation architectures. Pseudo control hedging (PCH) is used to protect the adaptive element from actuator saturation nonlinearities and also from the inner-outer-loop interaction. In addition, to augment the attitude control system, two online adaptive architectures that employ a neural network are used. By tuning the neural network based on the system model, a better and faster learning will be achieved, but this is a frustrating and time consuming process. Due to complexity in accurate tuning of neural network, this paper introduces a non-dominated sorting genetic algorithm II (NSGA-II) for off-line optimization of the neural network. Thus, in the proposed method, the neural network can compensate model inversion error caused by the deficiency of full knowledge of helicopter dynamics more accurately. The effectiveness of proposed method is demonstrated by numerical simulations.     

Selective ensemble extreme learning machine modeling of effluent quality in wastewater treatment plants

Real-time and reliable measurements of the effluent quality are essential to improve operating efficiency and reduce energy consumption for the wastewater treatment process.Due to the low accuracy and unstable performance of the traditional effluent quality measurements,we propose a selective ensemble extreme learning machine modeling method to enhance the effluent quality predictions.Extreme learning machine algorithm is inserted into a selective ensemble frame as the component model since it runs much faster and provides better generalization performance than other popular learning algorithms.Ensemble extreme learning machine models overcome variations in different trials of simulations for single model.Selective ensemble based on genetic algorithm is used to further exclude some bad components from all the available ensembles in order to reduce the computation complexity and improve the generalization performance.The proposed method is verified with the data from an industrial wastewater treatment plant,located in Shenyang,China.Experimental results show that the proposed method has relatively stronger generalization and higher accuracy than partial least square,neural network partial least square,single extreme learning machine and ensemble extreme learning machine model.     

Design of robotic visual servo control based on neural network and genetic algorithm

A new visual servo control scheme for a robotic manipulator is presented in this paper, where a back propagation (BP) neural network is used to make a direct transition from image feature to joint angles without requiring robot kinematics and camera calibration. To speed up the convergence and avoid local minimum of the neural network, this paper uses a genetic algorithm to find the optimal initial weights and thresholds and then uses the BP algorithm to train the neural network according to the data given. The proposed method can effectively combine the good global searching ability of genetic algorithms with the accurate local searching feature of BP neural network. The Simulink model for PUMA560 robot visual servo system based on the improved BP neural network is built with the Robotics Toolbox of Matlab. The simulation results indicate that the proposed method can accelerate convergence of the image errors and provide a simple and effective way of robot control.     

Gas recognition algorithm applied in porphyrin array gas detection system北大核心CSCD

The difference map is the basis of identifying gases by the PSA chips. However, there are differences between each difference map of a gas, which is called the“divergent problem”. A pattern recognition algorithm based on backpropagation neural network and rough set was described, which was employed in the porphyrin chemical sensor array integrated system. That algorithm picked up the spots whose color changed obviously using the rough set, and set their values as input of BP network. Comparing with the result of Euclidean distance clustering and BP neural network identification without removing unnecessary data as input, the result of the algorithm proposed in this article has higher identification accuracy to the divergence experimental data. ©, 2014, The Editorial Office of Chinese Journal of Sensors and Actuators. All right reserved.     

Nonlinear system PID-type multi-step predictive control

A compound neural network was constructed during the process of identification and multi-step prediction. Under the PID-type long-range predictive cost function, the control signal was calculated based on gradient algorithm. The nonlinear controller‘ s structure was similar to the conventional PID controller. The parameters of this controller were tuned by using a local recurrent neural network on-line. The controller has a better effect than the conventional PID controller. Sinmlation study shows the effectiveness and good performance.     

A fast learning algorithm of neural network with tunable activation function

This paper presents a modified structure of a neural network with tunable activation function and provides a new learning algorithm for the neural network training. Simulation results of XOR problem, Feigenbaum function, and Henon map show that the new algorithm has better performance than BP (back propagation) algorithm in terms of shorter convergence time and higher convergence accuracy. Further modifications of the structure of the neural network with the faster learning algorithm demonstrate simpler structure with even faster convergence speed and better convergence accuracy.     

An improved differential evolution trained neural network scheme for nonlinear system identification

This paper presents an improved nonlinear system identification scheme using di？erential evolution （DE）, neural network （NN） and Levenberg Marquardt algorithm （LM）. With a view to achieve better convergence of NN weights optimization during the training, the DE and LM are used in a combined framework to train the NN. We present the convergence analysis of the DE and demonstrate the efficacy of the proposed improved system identification algorithm by exploiting the combined DE and LM training of the NN and suitably implementing it together with other system identification methods, namely NN and DE＋NN on a number of examples including a practical case study. The identification results obtained through a series of simulation studies of these methods on different nonlinear systems demonstrate that the proposed DE and LM trained NN approach to nonlinear system identification can yield better identification results in terms of time of convergence and less identification error.     

基于神经网络集成的肺癌早期诊断

将病理性诊断与计算机技术相结合以实现肺癌的早期诊断，首先利用数字图像技术对肺癌穿刺样本进行处理，提出取形态和色度特征，然后通过一种二级集成结构和特殊的投票方式，用神经网络集成对细胞图象进行分析，实验和原型系统试用表明，方法的总误诊率和肺癌患者漏诊率均低于单一神经网络方法和常用的神经网络集成方法。     

Medical image diagnosis of lung cancer by a revised GMDH-type neural network using various kinds of neuron

A revised group method of data handling (GMDH)-type neural network algorithm using various kinds of neuron is applied to the medical image diagnosis of lung cancer. The optimum neural network architecture for medical image diagnosis is automatically organized using a revised GMDH-type neural network algorithm, and the regions of lung cancer are recognized and extracted accurately. In this revised GMDH-type neural network algorithm, polynomial-type and radial basis function (RBF)-type neurons are used for organizing the neural network architecture in order to fit the complexity of the nonlinear system.     

一种基于神经网络集成的决策树构造方法

神经网络集成方法具有比单个神经网络更强的泛化能力，却因为其黑箱性而难以理解；决策树算法因为分类结果显示为树型结构而具有良好的可理解性，泛化能力却比不上神经网络集成。该文将这两种算法相结合，提出一种决策树的构造算法：使用神经网络集成来预处理训练样本，使用C4．5算法处理预处理后的样本并生成决策树。该文在UCI数据上比较了神经网络集成方法、决策树C4．5算法和该文算法，实验表明：该算法具有神经网络集成方法的强泛化能力的优点，其泛化能力明显优于C4．5算法；该算法的最终结果昆示为决策树，显然具有良好的可理解性。     

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.     

基于卟啉传感器阵列系统的肺癌标志物识别算法

卟啉传感器阵列系统可以检测肺癌呼出气体中特定的标志性气体,不同标志性气体检测输出的差值图谱不一样.介绍了一种结合反向传播(BP)神经网络和主成分分析(PCA)的肺癌标志性气体种类识别算法,并将其应用在卟啉传感器阵列系统中.通过计算卟啉传感器阵列中各点的主成分得分选出敏感点,保留各气体敏感点的值,并组成识别模板作为BP神经网络的输入层,达到去除冗余数据的目的.通过实验对比聚类分析结果、未降维数据的BP神经网络识别结果及已经PCA降维后的数据作为输入的BP神经网络识别结果,证明提出的算法可以更加精确地识别不同的肺癌标志性气体.     

基于神经网络的肺癌细胞识别方法的研究

细胞识别是图像处理和模式识别领域的一个研究热点,有着十分广泛的应用前景。本文提出了基于神经网络算法FTART2的肺癌细胞识别方法,讨论了FTART2的网络结构、输入矢量的标准化及分类算法。用513个样本对网络进行训练,再用716个样本组成测试集进行测试,实验结果表明:本文提出的基于FTART2的肺癌细胞分类器与基于标准BP的分类器相比,具有学习速度快、分类精度高的特点。     

基于神经网络架构搜索的肺结节分类算法

肺结节分类是早期肺癌诊断的重要任务。基于深度学习的肺结节分类方法虽然能够取得良好的分类精度,但存在模型复杂和可解释性差的问题。为此,提出了一种基于神经网络架构搜索的肺结节分类算法。首先,将注意力残差卷积cell作为搜索空间的基本单元,并使用偏序剪枝方法作为搜索策略来构建神经网络架构以搜索3D分类网络,从而达到网络性能和搜索速度的平衡。其次,在网络中构建了多尺度通道和空间注意力模块来提高特征描述和类别推理的可解释性。最后,采用堆叠法将搜索到的网络架构进行多模型的融合,从而获取精准的肺结节良恶性分类预测结果。实验结果表明,在肺结节分类常用数据集LIDC-IDRI上,所提算法与最新肺结节分类算法相比具有较好的分类性能和较快的收敛,且所提算法的特异性和精确率分别达到95.37%和93.42%,能够实现良恶性肺结节的准确分类。     

最小一乘回归神经网络集成方法股市建模研究

提出了一种新的神经网络集成股市建模方法,采用偏最小二乘方法构造神经网络输入矩阵,利用Bagging技术和不同的神经网络学习算法生成集成个体,再用遗传算法选择参与集成的个体,以"误差绝对值和最小"为最优准,建立最小一乘回归神经网络集成模型,通过上证指数开盘价、收盘价进行实例分析,计算结果表明该方法具有较好的学习能力和泛化能力,在股市预测中预测精度高、稳定性好.     

一种基于神经网络集成的规则学习算法

将神经网络集成与规则学习相结合，提出了一种基于神经网络集成的规则学习算法．该算法以神经网络集成作为规则学习的前端，利用其产生出规则学习所用的数据集，在此基础上进行规则学习．在UCl机器学习数据库上的实验结果表明，该算法可以产生泛化能力非常强的规则．