基于K-means聚类和特征空间增强的噪声标签深度学习算法

深度学习中神经网络的性能依赖于高质量的样本,然而噪声标签会降低网络的分类准确率。为降低噪声标签对网络性能的影响,噪声标签学习算法被提出。该算法首先将训练样本集划分成干净样本集和噪声样本集,然后使用半监督学习算法对噪声样本集赋予伪标签。然而,错误的伪标签以及训练样本数量不足的问题仍然限制着噪声标签学习算法性能的提升。为解决上述问题,提出基于K-means聚类和特征空间增强的噪声标签深度学习算法。首先,该算法利用K-means聚类算法对干净样本集进行标签聚类,并根据噪声样本集与聚类中心的距离大小筛选出难以分类的噪声样本,以提高训练样本的质量;其次,使用mixup算法扩充干净样本集和噪声样本集,以增加训练样本的数量;最后,采用特征空间增强算法抑制mixup算法新生成的噪声样本,从而提高网络的分类准确率。并在CIFAR10、CIFAR100、MNIST和ANIMAL-10共4个数据集上试验验证了该算法的有效性。

基于特征分箱和K-Means算法的用户行为分析方法

针对网购用户所产生的购物行为进行分析,首先通过数据处理构建客户关系管理模型(RFM模型),在此模型的基础上采用特征分箱法和K-Means聚类两种方法对用户进行细分,并对2种模型结果进行比较分析,讨论二者的差异性和具体的应用范围和意义.其中,基于特征分箱法的RFM模型将变量转化到相似的尺度上并将变量离散化,使得用户分类标签更加清晰,也可依据各类标签分类出不同类型的用户.K-Means算法通过轮廓系数评估聚类算法质量以至于选取最优K值.本文实验分析结果可为运营商提供更加可靠直观的数据,使得运营商可以根据不同用户的不同行为进行市场细分,进而进行精准营销和服务设置.

Crack Fault Classification for Planetary Gearbox Based on Feature Selection Technique and K-means Clustering Method

During the condition monitoring of a planetary gearbox, features are extracted from raw data for a fault diagnosis.However, different features have different sensitivity for identifying different fault types, and thus, the selection of a sensitive feature subset from an entire feature set and retaining as much of the class discriminatory information as possible has a directly effect on the accuracy of the classification results. In this paper, an improved hybrid feature selection technique(IHFST) that combines a distance evaluation technique(DET), Pearson’s correlation analysis, and an ad hoc technique is proposed. In IHFST, a temporary feature subset without irrelevant features is first selected according to the distance evaluation criterion of DET, and the Pearson’s correlation analysis and ad hoc technique are then employed to find and remove redundant features in the temporary feature subset, respectively, and hence,a sensitive feature subset without irrelevant or redundant features is selected from the entire feature set. Further, the k-means clustering method is applied to classify the different kinds of health conditions. The effectiveness of the proposed method was validated through several experiments carried out on a planetary gearbox with incipient cracks seeded in the tooth root of the sun gear, planet gear, and ring gear. The results show that the proposed method can successfully distinguish the different health conditions of a planetary gearbox, and achieves a better classification performance than other methods. This study proposes a sensitive feature subset selection method that achieves an obvious improvement in terms of the accuracy of the fault classification.

结合人工蜂群与K-means聚类的特征选择

K-means聚类是一种简捷高效、收敛速度快且易于实现的统计分析方法,但是传统的K-means聚类算法对初始聚类中心的选取敏感且易陷入局部最优,同时多数无监督特征选择算法容易忽视特征之间的联系。为此,提出了一种结合人工蜂群与K-means聚类的特征选择方法。首先,为了使同一簇中样本的相似度高而不同簇中样本的相似度低,基于簇内聚集度和簇间离散度构建了新的适应度函数,更好地反映各样本的特性,进而构建了蜜源被选择新的概率表达式;其次,设计了随着迭代次数的增加而数值逐渐减小的权重,提出了使蜂群搜索范围动态缩进的蜜源位置更新表达式;然后,为了弥补传统的欧氏距离在计算距离时仅考虑向量之间的累积差异而表现出的局限性,构造了同时考虑样本影响程度不同以及样本的相似性的加权欧氏距离表达式;最后,引入标准差和距离相关系数,定义了特征区分度与特征代表性,以二者之积度量特征重要性。实验结果表明,所提算法加快了人工蜂群算法的收敛速度并提高了K-means算法的聚类效果,同时也有效地提升了特征选择的分类效果。

基于BERT-LDA和K-means聚类的绘画作品价值评估指标体系构建

针对目前绘画领域缺乏标准的价值评估指标体系,提出了基于BERT-LDA和K-means聚类的绘画作品价值要素挖掘方法。运用超平面法对绘画文献进行了停用词筛选,基于BERT-LDA模型构建了包含文本语义信息的融合特征向量,运用K-means算法对融合特征向量进行降维可视化,随之构建了绘画作品价值评估指标体系。结果表明,基于BERT-LDA模型和K-means算法识别的主题及主题词相比传统LDA模型的查准率、查全率和F值分别提升了28.5%、10%和21.5%。通过随机森林等算法对指标体系进行验证,验证了构建的绘画作品价值评估指标体系的科学性。

基于改进K-means数据聚类算法的网络入侵检测

随着入侵手段的不断更新和升级,传统入侵检测方法准确率下降、检测时间延长,无法满足网络防御要求。为此,提出一种经过改进K均值(K-means)数据聚类算法,以应对不断升级的网络入侵行为。先以防火墙日志为基础转换数值,然后基于粒子群算法求取最优初始聚类中心,实现K-means数据聚类算法的改进;最后以计算得出的特征值为输入项,实现对网络入侵行为的精准检测。结果表明:K-means算法改进后较改进前的戴维森堡丁指数更小,均低于0.6,达到了改进目的。改进K-means算法各样本的准确率均高于90%,相对更高,检测时间均低于10 s,相对更少,说明该方法能够以高效率完成更准确的网络入侵检测。

基于自适应布谷鸟优化特征选择的K-means聚类

K-means聚类算法随机确定初始聚类数目,而且原始数据集中含有大量的冗余特征会导致聚类时精度降低,而布谷鸟搜索(CS)算法存在收敛速度慢和局部搜索能力弱等问题,为此提出一种基于自适应布谷鸟优化特征选择的K-means聚类算法(DCFSK)。首先,为提升CS算法的搜索速度和精度,在莱维飞行阶段,设计了自适应步长因子;为调节CS算法全局搜索和局部搜索之间的平衡、加快CS算法的收敛,动态调整发现概率,进而提出改进的动态CS算法(IDCS),在IDCS的基础上构建了结合动态CS的特征选择算法(DCFS)。其次,为提升传统欧氏距离的计算精确度,设计同时考虑样本和特征对距离计算贡献程度的加权欧氏距离;为了确定最佳聚类数目的选取方法,依据改进的加权欧氏距离构造了加权簇内距离和簇间距离。最后,为克服传统K-means聚类目标函数仅考虑簇内的距离而未考虑簇间距离的缺陷,提出基于中位数的轮廓系数的目标函数,进而设计了DCFSK。实验结果表明,在10个基准测试函数上,IDCS的各项指标取得了较优的结果;相较于K-means、DBSCAN(Density-Based Spatial Clustering of Applications with Noise)等算法,在6个合成数据集与6个UCI数据集上,DCFSK的聚类效果最佳。

融合k-means与多维特征分析的生物医学图像分类算法

为了提高生物医学图像分类的准确率与速度,提升医学研究者及临床工作者的工作效率,文中提出了一种融合k-means与多维特征分析的生物医学图像分类算法。该算法利用局部二值模式LBP与方向梯度直方图HOG分别提取医学图像中的纹理特征及局部特征两种不同维度的特征信息,并将k-means聚类算法与这种多维特征分析相融合,实现了对生物医学图像的高精度分类。在公开生物医学图像数据集BreakHis上进行的仿真实验结果表明,在二分类实验中,所提算法的准确率为99.03%、精确率为99.12%、召回率为98.96%、F1值为99.04%,在八分类实验中其性能也较为理想,优于SVM、ELM及ResNet等分类算法。

一种基于K-means的神经网络数据集回归预测算法

在智能电网研究领域的高维数据回归分析和预测模型中,传统的统计学模型不能平衡不同维度之间信息价值,影响数据集的预测有效性。为解决上述问题,提出一种基于K-means的神经网络数据集回归预测算法。首先,在特征层面上,多层循环神经网络提取不同维度的数据特征并训练响应,然后,在算法层面上,通过K-means的分类器模型依照数据的维度特征分类并融合循环神经网络(Recurrent Neural Network,RNN)的特征响应,再对输出响应的数据集构建组合预测模型,从而提高预测算法的可靠性。在公开的回归数据集上进行测试。实验测试的结果表明,与门控循环算法(Gated Recurrent Unit,GRU)相比降低了14.45%的平均绝对误差值。

A Hybrid K-Means-GRA-SVR Model Based on Feature Selection for Day-Ahead Prediction of Photovoltaic Power Generation

In order to ensure that the large-scale application of photovoltaic power generation does not affect the stability of the grid, accurate photovoltaic (PV) power generation forecast is essential. A short-term PV power generation forecast method using the combination of K-means++, grey relational analysis (GRA) and support vector regression (SVR) based on feature selection (Hybrid Kmeans-GRA-SVR, HKGSVR) was proposed. The historical power data were clustered through the multi-index K-means++ algorithm and divided into ideal and non-ideal weather. The GRA algorithm was used to match the similar day and the nearest neighbor similar day of the prediction day. And selected appropriate input features for different weather types to train the SVR model. Under ideal weather, the average values of MAE, RMSE and R2 were 0.8101, 0.9608 kW and 99.66%, respectively. And this method reduced the average training time by 77.27% compared with the standard SVR model. Under non-ideal weather conditions, the average values of MAE, RMSE and R2 were 1.8337, 2.1379 kW and 98.47%, respectively. And this method reduced the average training time of the standard SVR model by 98.07%. The experimental results show that the prediction accuracy of the proposed model is significantly improved compared to the other five models, which verify the effectiveness of the method.

基于Bag of Features算法的车辆图像识别研究

将Bag of Features算法引入汽车图像识别领域中,并提出了将DoG(Difference of Gaussian)特征提取算法和PLSA分类算法结合在一起实现车辆和背景图像分类。首先用DoG特征提取算法提取图像特征,用这些特征聚类产生码书并对图像进行柱状图描述,最后设计PLSA分类器对车辆图像和背景图像进行分类。实验对比了该算法与Tamura纹理特征算法和Gabor纹理特征算法在车辆图像识别中的效果。结果表明本文算法分类正确率优于另外两种方法。

Feature Patch Illumination Spaces and Karcher Compression for Face Recognition via Grassmannians

Recent work has established that digital images of a human face, when collected with a fixed pose but under a variety of illumination conditions, possess discriminatory information that can be used in classification. In this paper we perform classification on Grassmannians to demonstrate that sufficient discriminatory information persists in feature patch (e.g., nose or eye patch) illumination spaces. We further employ the use of Karcher mean on the Grassmannians to demonstrate that this compressed representation can accelerate computations with relatively minor sacrifice on performance. The combination of these two ideas introduces a novel perspective in performing face recognition.

Genetic Feature Selection for Texture Classification

This paper presents a novel approach to feature subset selection using genetic algorithms. This approach has the ability to accommodate multiple criteria such as the accuracy and cost of classification into the process of feature selection and finds the effective feature subset for texture classification. On the basis of the effective feature subset selected, a method is described to extract the objects which are higher than their surroundings, such as trees or forest, in the color aerial images. The methodology presented in this paper is illustrated by its application to the problem of trees extraction from aerial images.

基于K-means与KNN的多特征苹果在线分级

【目的】准确获取红富士苹果的分级指标,为实现多特征融合的苹果分级提供依据。【方法】以均值滤波、全局亮度均衡化与图像裁剪方法,预处理实验所需的苹果图像;使用K-means聚类算法、OTSU最大类间方差法,将苹果灰度图转换为二值图;利用二值图与苹果原图的异或运算,提取苹果轮廓;采用苹果的二值图计算苹果的果实区域大小;使用颜色空间转换RGB-HSV中H通道划分果实红色区域;通过构建掩膜、形态学操作判断果体是否含有缺陷及计算其面积;构建最小外接矩形计算苹果的果径及果形;利用KNN分类算法实现多特征融合的苹果在线自动分级。【结果】基于K-means聚类与KNN分类相结合的苹果在线分级方法,在优于传统图像阈值分割效果的基础上,特级果分级准确率为97.14%,一级果分级准确率为100%,二级果分级准确率为93.75%,等外果分级准确率为100%,综合分级准确率达到97%。【结论】100个苹果测试准确率达到97%,验证了该分级方法的可行性与准确性。

基于K-means聚类和改进MLP的苹果分级研究

为准确实现多特征融合的苹果分级,提出了一种基于K-means聚类和改进MLP的苹果分级方法。该方法主要包括图像预处理、亮度均衡化、背景分割、特征加权以及改进的MLP分级网络训练。首先借助均值滤波算法和直方图均衡化操作改善苹果图像质量;接着借助K-means聚类算法进行背景分割;在果体与背景分割的基础上,依次提取苹果的果径、果形、颜色、缺陷、纹理5个特征;然后借助皮尔逊相关性分析和人工挑选偏好权重对特征数据集综合加权,模拟人工分级场景;最后将特征数据送入改进的MLP神经网络中完成苹果的分级定等。通过对400个定好等级的苹果进行分级测试,准确率达到94.25%,验证了分级方法的可行性与准确性。该方法与现行的苹果分级标准相结合,具备时效性强、检测指标完备等分级优势。

Adaptive WNN aerodynamic modeling based on subset KPCA feature extraction

In order to accurately describe the dynamic characteristics of flight vehicles through aerodynamic modeling, an adaptive wavelet neural network (AWNN) aerodynamic modeling method is proposed, based on subset kernel principal components analysis (SKPCA) feature extraction. Firstly, by fuzzy C-means clustering, some samples are selected from the training sample set to constitute a sample subset. Then, the obtained samples subset is used to execute SKPCA for extracting basic features of the training samples. Finally, using the extracted basic features, the AWNN aerodynamic model is established. The experimental results show that, in 50 times repetitive modeling, the modeling ability of the method proposed is better than that of other six methods. It only needs about half the modeling time of KPCA-AWNN under a close prediction accuracy, and can easily determine the model parameters. This enables it to be effective and feasible to construct the aerodynamic modeling for flight vehicles.

Different Feature Selection of Soil Attributes Influenced Clustering Performance on Soil Datasets

Feature selection is very important to obtain meaningful and interpretive clustering results from a clustering analysis. In the application of soil data clustering, there is a lack of good understanding of the response of clustering performance to different features subsets. In the present paper, we analyzed the performance differences between k-means, fuzzy c-means, and spectral clustering algorithms in the conditions of different feature subsets of soil data sets. The experimental results demonstrated that the performances of spectral clustering algorithm were generally better than those of k-means and fuzzy c-means with different features subsets. The feature subsets containing environmental attributes helped to improve clustering performances better than those having spatial attributes and produced more accurate and meaningful clustering results. Our results demonstrated that combination of spectral clustering algorithm with the feature subsets containing environmental attributes rather than spatial attributes may be a better choice in applications of soil data clustering.

基于k-means聚类算法与多维特征融合的群体划分模型

常规群体划分模型构建方法主要采用二分均值算法,该方法易受到数据特征项贡献度的影响,使得模型的划分结果准确性较低。因此,提出基于k-means聚类算法与多维度特征融合的群体划分模型。利用k-means聚类算法通过合理设定密度阈值与邻域半径,提取用户群体的兴趣区域,在兴趣区域中选取贡献度较大的数据特征项,并计算特征项的权值,以此为依据,采用多维特征融合算法改进最大化目标函数,以此实现群体划分模型的构建。实验结果显示,利用所提方法构建的群体划分模型,能够得到较高的划分准确度。

The Conversion of Nouns to Verbs in English

In Chinese Ianguage, there are many nouns used as verbs. In fact, there is a same phenomenon in English. The paper includes; the preponderance of nouns over verbs, the classification and development of nouns to verbs conversion, the three features of nouns to verbs conversion.