Feature selection considering weighted relevancy

Feature selection plays an important role in pattern recognition and machine learning. Feature selection based on information theory intends to preserve the feature relevancy between features and class labels while eliminating irrelevant and redundant features. Previous feature selection methods have offered various explanations for feature relevancy, but they ignored the relationships between candidate feature relevancy and selected feature relevancy. To fill this gap, we propose a feature selection method named Feature Selection based on Weighted Relevancy (WRFS). In WRFS, we introduce two weight coefficients that use mutual information and joint mutual information to balance the importance between the two kinds of feature relevancy terms. To evaluate the classification performance of our method, WRFS is compared to three competing feature selection methods and three state-of-the-art methods by two different classifiers on 18 benchmark data sets. The experimental results indicate that WRFS outperforms the other baselines in terms of the classification accuracy, AUC and F1 score.     

PolSAR ship detection based on the contrast enhancement of polarimetric scattering differences

The ship detection in polarimetric synthetic aperture radar (PolSAR) mode is a hot topic in recent years, because of the diversity of polarimetric scattering mechanisms between ship targets and sea clutter. To improve the detection performance of ship targets, this paper mainly develops the ship detection method based on the contrast enhancement utilizing the polarimetric scattering difference. The algorithm first enhances the target signal utilizing the scattering difference of the polarimetric coherency matrix between ship targets and sea clutter, and then a simple threshold is applied to distinguish the ship targets from the sea clutter. Finally, real PolSAR datasets recorded by AirSAR system are used to evaluate the effectiveness of the proposed detection method. Compared with other detection methods, experimental results indicate that the proposed method can effectively improve the detection performance of ship targets.     

基于改进CHI和带权ECE结合的特征选择方法

针对文本分类特征选择方法中的卡方统计（CHI）和期望交叉熵（ECE）,分析了其特点和不足。为了避免传统CHI和ECE方法在不平衡数据集上分类效果差的问题,通过引入调节因子和除去负相关影响因素,给出了改进的CHI方法（◢p◣CHI）,并以加权的方式弥补ECE方法倾向于选择弱区分能力高频特征的缺陷（◢ω◣ECE）。在综合两种改进后方法的基础上,进一步提出基于改进CHI和带权ECE结合（◢p◣CHI◢ω◣ECE）的特征选择方法。经对比实验验证,◢p◣CHI◢ω◣ECE方法的查准率、◢F◣▼1▽值均优于CHI、ECE及◢p◣CHI、◢ω◣ECE方法,且该方法的降维稳定性更好。     

Feature selection via sensitivity analysis of SVM probabilistic outputs

Feature selection is an important aspect of solving data-mining and machine-learning problems. This paper proposes a feature-selection method for the Support Vector Machine (SVM) learning. Like most feature-selection methods, the proposed method ranks all features in decreasing order of importance so that more relevant features can be identified. It uses a novel criterion based on the probabilistic outputs of SVM. This criterion, termed Feature-based Sensitivity of Posterior Probabilities (FSPP), evaluates the importance of a specific feature by computing the aggregate value, over the feature space, of the absolute difference of the probabilistic outputs of SVM with and without the feature. The exact form of this criterion is not easily computable and approximation is needed. Four approximations, FSPP1-FSPP4, are proposed for this purpose. The first two approximations evaluate the criterion by randomly permuting the values of the feature among samples of the training data. They differ in their choices of the mapping function from standard SVM output to its probabilistic output: FSPP1 uses a simple threshold function while FSPP2 uses a sigmoid function. The second two directly approximate the criterion but differ in the smoothness assumptions of criterion with respect to the features. The performance of these approximations, used in an overall feature-selection scheme, is then evaluated on various artificial problems and real-world problems, including datasets from the recent Neural Information Processing Systems (NIPS) feature selection competition. FSPP1-3 show good performance consistently with FSPP2 being the best overall by a slight margin. The performance of FSPP2 is competitive with some of the best performing feature-selection methods in the literature on the datasets that we have tested. Its associated computations are modest and hence it is suitable as a feature-selection method for SVM applications. Editor: Risto Miikkulainen.     

特征选择的轻量级入侵检测系统

特征选择能够很好地消除冗余和噪音特征,有利于提高入侵检测系统的检测速度和效果,因而对基于特征选择的入侵检测系统进行研究是必要的,也符合入侵检测领域的发展趋势。提出了一种基于过滤器模式的轻量级入侵检测系统,无论是在数据集的特征选择算法还是分类器的参数优化上,都给出了有效的实施策略,提高了检测速度,降低了分类干扰,提高了入侵检测的检测率。     

Damage assessment in urban areas using post-earthquake airborne PolSAR imagery

Synthetic aperture radar (SAR) has often been used in earthquake damage assessment due to its extreme versatility and almost all-weather, day-and-night capability. In this article, we demonstrate the potential to use only post-event, high-resolution airborne polarimetric SAR (PolSAR) imagery to estimate the damage level at the block scale. Intact buildings with large orientation angles have a similar scattering mechanism to collapsed buildings; they are all volume-scattering dominant and reflection asymmetric, which seriously hampers the process of damage assessment. In this article, we propose a new damage assessment method combining polarimetric and spatial texture information to eliminate this deficiency. In the proposed method, the normalized circular-pol correlation coefficient is used first to identify intact buildings aligned parallel with the flight direction of the radar. The ‘homogeneity’ feature of the grey-level co-occurrence matrix (GLCM) is then introduced to distinguish building patches with large orientation angles from the severely damaged class. Furthermore, a new damage assessment index is also introduced to handle the assessment at the level of the block scale. To demonstrate the effectiveness of the proposed approach, the high-resolution airborne PolSAR imagery acquired after the earthquake that hit Yushu County, Qinghai Province of China, is investigated. By comparison with the damage validation map, the results confirm the validity of the proposed method and the advantage of further improving the assessment accuracy without external ancillary optical or SAR data.     

多通道三维视觉指导运动想象脑电信号特征选择算法

针对基于三维视觉指导的运动想象脑机接口多通道冗余信息较多、分类准确率差的问题,提出了一种基于小波包分解（WPD）—共空间滤波（CSP）—自适应差分进化（ADE）的模式脑电信号特征提取与选择分类方法。首先,对采集的多通道运动想象脑电信号进行WPD变化,划分出精细的子频带;然后,分别将WPD变换后的每个子空间作为CSP的输入,得到对应的特征向量;最后,使用ADE算法对特征向量进行选择,选择出用于分类的最佳特征子集。采用WPD-CSP-ADE模式进行特征提取与选择,较经典的WPD-CSP方法在分类正确率、特征个数方面有着更好的表现。同时,所提算法分类性能明显优于遗传算法、粒子群算法。实验结果表明,WPD-CSP-ADE方法能够有效地提高分类正确率,同时减少了用于分类的特征个数。     

A weighted intrusion detection model of dynamic selection

In view of the difficulty of existing intrusion detection methods in dealing with new forms, large scale, and high concealment of network intrusion behaviors, this paper presents a weighted intrusion detection model of the dynamic selection (WIDMoDS) based on data features. The aim is to customize intrusion detection models for network intrusion data sets of different types, sizes and structures. First, according to data features, single classifiers are clustered using a hierarchical clustering algorithm based on the classifiers evaluation indicators, and then, the classifiers selection is by means of accuracy of the single classifiers, in addition, the data-classifier applicable indicators (DCAI) and of the classifiers performances are used for calculating the weights of subjective and objective, and then calculating combined weight ranks. Finally, a custom intrusion detection model is generated by the Weight-voting (W-voting) algorithm. Our experiments show that this model can optimize the number of classifiers based on the data sets features, reduce the problem of redundant or insufficient classifiers in the ensemble process. A new network intrusion detection model of combining the classifier characteristics with the dataset attributes can improve the accuracy of intrusion detection.

     

异常流量检测中的特征选择

针对异常流量检测领域的高维数据降维问题,提出了一种基于信息熵理论的特征选择算法。首先计算特征的重要系数,删除重要系数小于一定阈值的特征,得到重要特征集。然后,计算特征间的冗余系数,删除冗余特征,得到精简的特征集。最后,用ID3算法对精简的特征集进行了验证,结果表明这种特征选择算法是有效的。     

Enhanced duckweed detection using bootstrapped SVM classification on medium resolution RGB MODIS imagery

From early 2004, Lake Maracaibo (northwest Venezuela) experienced an unprecedented invasion of duckweed Lemna obscura. Recurrent blooms of the plant in the past 2 years illustrate the need for an automatic monitoring method to follow the plant cover with time and to plan contingency measures. We present an approach that allows the cover of the duckweed to be quantified through the classification of MODIS 250 m RGB composite images available from the internet. The method improves the accuracy of the results of the Support Vector Machine (SVM) algorithm for classification by including a bootstrap step during the training phase. Using only 200 pixels for training (<0.05% of the total), the bootstrapped SVM method allows a better identification of the duckweed class, reducing the number of false negatives by half and improving the KHAT statistic by almost 40% in comparison to the standard SVM method. This method has proved to be a reliable solution in cases where rapid responses are needed and only medium‐resolution, free satellite imagery is available.     

基于带特征染色体遗传算法的支持向量机特征选择和参数优化

鉴于支持向量机特征选择和参数优化对其分类准确率有重大的影响,将支持向量机渐近性能融入遗传算法并生成特征染色体,从而将遗传算法的搜索导向超参数空间中的最佳化误差直线.在此基础上,提出一种新的基十带特征染色体遗传算法的方法,同时进行支持向量机特征选择和参数优化.在与网格搜索、不带特征染色体遗传算法和其他方法的比较中,所提出的方法具有较高的准确率、更小的特征子集和更少的处理时间.     

基于样本选取和加权KPCA-L1的异常检测*

为了提高基于一范数的核主成分分析算法（KPCA-L1）处理异常检测问题的速度,提出了基于样本选取和加权KPCA-L1的异常检测方法。所提方法首先从训练集中选取具有代表性的特征子集,然后为所得特征子集中的样本赋予权重,用带有权重的特征子集训练模型,构造加权KPCA-L1。与KPCA-L1相比,所提方法能够有效地减小训练集的规模,同时改善了KPCA-L1算法的更新方法。在人工数据集和标准数据集上的实验结果表明,在保证异常检测准确率的前提下,所提方法比KPCA-L1具有更快的建模速度。     

基于局部密度的加权一类支持向量机算法及其在涡轴发动机故障检测中的应用

针对基于数据的涡轴发动机故障检测算法的分类性能较差、鲁棒性不强的问题,提出一种改进的加权一类支持向量机（WOCSVM）算法——基于局部密度的WOCSVM （LD-WOCSVM）算法。首先,对于每个训练样本,选取以该样本为中心,以全体训练样本中心到距离最远样本之间马氏距离的百分之二为半径的球体内所包含的k个近邻样本;其次,以该样本到选定的k个训练样本的中心的距离大小来评估该样本为故障样本的可能性,并以此为依据,使用经过归一化的距离来计算对应样本的权重。针对目前算法不能很好地反映样本分布特点的问题,提出了一种基于快速聚类的权重计算方法并将其命名为FCLD-WOCSVM。该算法通过求取每个训练样本的局部密度和该样本到高局部密度的距离两个参数,来确定该样本的分布位置,并利用求得的两个参数来计算该样本的权重。两种算法都是通过对可能的故障样本分配较小的权重来增强算法的分类性能。为了验证算法的有效性,分别在4个UCI数据集和T700涡轴发动机上进行仿真实验。实验结果表明,与自适应WOCSVM （A-WOCSVM）算法相比,LD-WOCSVM算法在AUC值上提高了0.5%,FCLD-WOCSVM算法在G-mean上提高了12.1%,两种算法可以作为涡轴发动机故障检测候选算法。     

基于遗传算法的入侵检测特征选择*

针对入侵检测日志数据存在大量不相关特征和冗余特征,导致入侵检测数据集维数较高,检测算法实时性较低的问题,提出一种基于遗传算法的入侵检测特征选择算法。首先删除入侵检测数据集中的不相关特征及冗余特征,构建有效特征集L,并通过偏F检验对特征进一步选择,构成待优化特征集L’;然后采用遗传算法对L’进行优化选择,选出最能反映系统状态的特征集L″。仿真实验结果证明,该算法在保证特征分类精度和确保入侵检测漏检率、误检率尽量小的前提下明显提高了入侵检测的效率。     

An approach for visual attention based on biquaternion and its application for ship detection in multispectral imagery

This paper proposes an approach for visual attention based on biquaternion, and investigates its application for ship detection in multispectral imagery. The proposed approach describes high-dimensional data in the form of biquaternion and utilizes the phase spectrum of biquaternion Fourier transform to generate a required saliency map that can be used for salient target detection. In our method, the multidimensional data is processed as a whole, and the features contained in each spectral band can be extracted effectively. Compared with traditional visual attention approaches, our method has very low computational complexity. Experimental results on simulated and real multispectral remote sensing data have shown that the proposed method has excellent performance in ship detection. Furthermore, our method is robust against white noise and almost meets real-time requirements, which has great potentials in engineering applications.     

Improved segmentation and change detection of multi-spectral satellite imagery using graph cut based clustering and multiclass SVM

Multimedia Tools and Applications - The Satellite image analysis automatically looks over an image to attain valuable information such as land cover classification and change detection from it....     

基于多目标进化算法的入侵检测特征选择

针对入侵检测系统要求检测率和误报率均衡优化,提出一种由顺序搜索策略改进的多目标进化算法,对特征空间进行压缩,以选择最优特征子集。实验结果表明,改进的多目标进化算法实现了检测率与误报率的均衡优化,较好地提高了入侵检测系统的性能。     

一种聚类加权支持向量机算法及其在软测量中的应用

针对支持向量机应用于软测量建模时,工业过程数据中特异点影响建模精度的问题,提出聚类加权支持向量机方法.该方法首先对建模数据进行聚类分析,根据聚类结果,对各类数据的惩罚系数进行相应的加权,改变权值大小既能减小特异点对模型的影响程度,又能将其包含的生产过程信息引入到软测量模型中.聚丙烯熔融指数软测量的实例研究表明,通过对建模数据进行聚类分析和加权处理,聚类加权支持向量机比标准支持向量机建模更准确.     

Feature selection by genetic algorithms in object-based classification of IKONOS imagery for forest mapping in Flanders, Belgium

Obtaining detailed information about the amount of forest cover is an important issue for governmental policy and forest management. This paper presents a new approach to update the Flemish Forest Map using IKONOS imagery. The proposed method is a three-step object-oriented classification routine that involves the integration of 1) image segmentation, 2) feature selection by Genetic Algorithms (GAs) and 3) joint Neural Network (NN) based object-classification. The added value of feature selection and neural network combination is investigated. Results show that, with GA-feature selection, the mean classification accuracy (in terms of Kappa Index of Agreement) is significantly higher (p < 0.01) than without feature selection. On average, the summed output of 50 networks provided a significantly higher (p < 0.01) classification accuracy than the mean output of 50 individual networks. Finally, the proposed classification routine yields a significantly higher (p < 0.01) classification accuracy as compared with a strategy without feature selection and joint network output. In addition, the proposed method showed its potential when few training data were available.