基于基因功能表达谱的疾病分类:抗基因缺失的稳健性

利用基因表达谱数据,按Gene Ontology基因功能分类体系,将基因模块化地组织到具有显著生物学意义的低维差异表达功能模块单元中,构造新的指标用于分类疾病样本,从而提出了基于功能表达谱的分析新途径。新算法可稳健地抗基因检测缺失,抗基因表达变异,抗检测误差,并可以显著地降低分类特征维数(参与疾病分类的基因数目)。采用淋巴瘤数据集,比较了基于功能表达谱和常规的基因表达谱的决策树分类器。结果显示,基于功能表达谱可以得到高准确度的疾病样本分类结果,能够直接从功能水平上给出相应的生物学解释。通过仿真分析,进一步显示了基于功能表达谱的分类方法具有抗基因检测缺失的稳健性。     

融合夹角度量的局部线性嵌入算法

局部线性嵌入(LLE)等流形学习算法中需要通过欧氏距离来度量数据点之间的近邻关系,但欧氏距离只表示两点间的直线距离,在高维空间中不一定能真实反映出图像数据点之间的空间分布情况.针对此问题,本文提出了融合数据间夹角和欧氏距离度量LLE近邻和分类的方法.该方法通过融合图像数据间的夹角和欧氏距离来度量图像数据点之间的近邻关系,寻找k个近邻点,实现更有效的局部重构,提取鉴别特征,然后用融合了数据间夹角的最近邻分类器对数据进行分类.在KSC和Indian Pine高光谱遥感影像数据集上的实验结果表明:在总体分类精度上,本文算法比LLE提升了1.54%～6.91％.     

融合Gabor小波和双层LLE的人耳图像识别

针对局部线性嵌入(LLE)算法易受偏转角度和近邻选取的影响,提出了一种双层LLE(DLLE)的人耳图像识别方法,并结合Gabor小波和DLLE提出了GDLLE。DLLE首先计算各样本与每类样本中心的欧氏距离,再把欧氏距离最小的K类所有样本作为LLE的近邻,提取出鉴别特征,最后由最近近邻分类器对鉴别特征进行分类。在USTB3人耳图像库上的实验结果表明,本文提出DLLE能够减小偏转角度和近邻对LLE算法的影响,结合Gabor小波后进一步改善了算法的识别率。     

一种改进的邻域保持嵌入高光谱影像分类方法

为进一步提高邻域保持嵌入算法(NPE)在高光谱影像分类中的识别性能,提出一种改进的半监督邻域保持嵌入(SSNPE)算法。首先,该算法在NPE算法的基础上同时利用同类标记样本和邻域未标记样本获得数据的邻域嵌入结构。然后,通过增加近邻标记样本的权重加大降维数据的鉴别性。最后,通过利用k近邻分类器(KNN)对样本进行分类得到该算法在数据集上的分类性能。在Urban、Indian高光谱影像数据集上的实验结果表明,改进的算法的分类精度相比其他算法提高了约8.3%、6.2%以上,分类性能上有了较为明显的提高。     

异构分类器融合环境下的非平衡数据分类模型

为解决非平衡数据分类中的正样本分类精度不高的瓶颈问题,提出了一种异构分类器融合环境下的非平衡数据分类模型.该模型基于差异采样率的重采样算法和改进的Adaboost算法,融合了SVM和C5.0两种基分类器;基于知识融合机制,采用了独特的分类器选择策略、分类器集成方法、分类决策方案.仿真实验结果表明,SCECM模型分类性能...     

一种基于相关函数法的奇异值补值方法

基于平稳随机序列受自相关函数约束的特点,研究了对随机信号中奇异值或缺失值进行补值的方法．利用平稳随机序列的一步变化率与自相关函数变化率之间的关系,得到一种基于一步相关法的补值处理方法,并给出这种方法的实现步骤．仿真试验表明,该方法对孤立和椒盐（斑点或连续）型散布的奇异值或缺失值的补值处理具有很好的收敛性和处理精度．     

基于MRMHC-LSVM的IP流分类

提出了一种构建轻量级的IP流分类器的wrapper型特征选择算法MRMHC-LSVM.该算法采用改进的随机变异爬山(MRMHC)搜索策略对特征子集空间进行随机搜索,然后利用提供的数据在无约束优化线性支持向量机(LSVM)上的分类错误率作为特征子集的评价标准来获取最优特征子集.在IP流数据集上进行了大量的实验,实验结果表明基于MRMHC-LSVM的流分类器在不影响分类准确度的情况下能够提高检测速度,与当前典型的流分类器NBK-FCBF相比,基于MRMHC-LSVM的IP流分类器具有更小的计算复杂度与更高的检测率.     

PSO和Cholesky分解的KELM的基因表达数据分类

核极限学习机(KELM)可使低维空间中线性不可分的数据变得线性可分,增加了ELM算法的鲁棒性,但KELM算法的输入权值参数采用随机初始化,容易导致算法不稳定.为此,本研究提出用粒子群优化算法对KELM中的权值初始参数进行优化、设定,以得到优化的分类器PSO-KELM.由于该算法输出权值求解采用传统的矩阵求逆运算,导致计算复杂,因此再对KELM的输出权值采用Cholesky分解进行优化.经一些标准基因数据集的实验表明,提出的PSO-KELM算法与已有的ELM、KELM、PSO-ELM相比分类精度更高,适用于基因表达数据分类.     

基于集成分类器的泡罩包装药品缺陷识别

目的 针对药品生产包装过程中常出现缺陷泡罩包装药品的问题,研究一种基于多特征构建与集成分类器的泡罩包装药品缺陷识别方法.方法 该方法通过集成2个不同的分类器算法分别对药品图像类别进行预测,并采用联合判定函数对2个预测输出值进行联合决策,得到最终分类结果.第1个分类器模型通过将图像转化到HSV颜色空间,分割出泡罩区域和药片区域,进行特征设计,并在提取多项特征参数后构建BP神经网络分类算法给定药品类别预测.第2个分类器模型应用多层卷积神经网络取代传统算法对图像特征进行提取,并输出药品图像类别的预测值.根据2个分类器的性能进行算法集成,构成最终集成分类器.结果 实验结果表明,该集成分类模型对数据集中泡罩包装药品图像进行分类识别测试,准确率达97％以上.结论 集成分类模型不仅提高了单一分类器的识别准确率,也具有更佳的稳定性.该方法取得了卓越的分类效果,具有较高应用性.     

印刷套准识别方法研究

目的研究印刷标志套准机器快速和高精度的识别方法。方法提取印刷标志图像的灰度共生矩阵表达其纹理特征,采用Adaboost分类器对印刷标志套准图像进行识别,以判断印刷是否套准。结果提取出了印刷标志图像的能量、熵、惯性矩、相关度等的均值和标准差的8维图像纹理特征。为了比较不同分类器的分类性能,分别得出了Adaboost、K近邻、贝叶斯、支持向量机、Fisher和决策树对印刷标志图像纹理特征的分类准确率和分类时间。结论采用文中方法,印刷标志图像套准识别准确率达到97.5%,分类时间达到0.022 377 s,优于其他的分类方法。     

基于BPSO-KNN算法的被动声呐目标分类识别技术研究

以提取得到的被动声呐目标功率谱特征为基础,采用二进制粒子群(Binary Particle Swarm Optimization, BPSO)优化算法和k最近邻(k-Nearest Neighbor, KNN)分类算法相结合的BPSO-KNN算法进行特征选择和参数优化,分别用KNN分类算法和BPSO-KNN分类算法对实际得到的四类海上被动声呐目标进行分类识别。结果表明,BPSO-KNN算法可对提取的功率谱特征进行特征优化选择,并对KNN分类器进行参数优化,提高了对四类目标的分类精度。该算法在被动声呐目标分类识别方面有参考价值。     

基于字典学习的主动声呐目标分类方法

主动声呐目标分类在军事和民用方面都有重要的应用和价值。文章基于稀疏表示理论,结合K-奇异值分解和正交匹配追踪算法,提出一种基于学习字典的稀疏表示分类方法（Dictionary Learning Sparse Representation Classification,DLSRC）。首先,利用K-奇异值分解算法训练各个类别目标回波信号,得到带有目标特征信息的类别字典,类别字典对信号具有良好表征能力并且带有目标类别信息;然后,利用正交匹配追踪算法和各个类别字典稀疏分解测试信号,得到各个类别字典下的稀疏系数后重构信号;最后,根据各个重构信号与测试信号的匹配度判定类别,得到分类准确率。结果显示,200个测试数据在信噪比分别为-5、-3、6 dB时,DLSRC法的分类准确率分别达到87%、89%、95.5%。不同信噪比下基于学习字典稀疏表示分类方法的准确率均高于已有的支持向量机（Support Vector Machine,SVM）、K-最近邻（K-Nearest Neighbor,KNN）和柔性最大值分类器（SoftMax）等分类方法,具有较好的分类性能。     

An Intelligent Gestational Diabetes Diagnosis Model Using Deep Stacked Autoencoder

Gestational Diabetes Mellitus (GDM) is one of the commonly occurring diseases among women during pregnancy. Oral Glucose Tolerance Test (OGTT) is followed universally in the diagnosis of GDM diagnosis at early pregnancy which is costly and ineffective. So, there is a need to design an effective and automated GDM diagnosis and classification model. The recent developments in the field of Deep Learning (DL) are useful in diagnosing different diseases. In this view, the current research article presents a new outlier detection with deep-stacked Autoencoder (OD-DSAE) model for GDM diagnosis and classification. The goal of the proposed OD-DSAE model is to find out those mothers with high risks and make them undergo earlier diagnosis, monitoring, and treatment compared to low-risk women. The presented OD-DSAE model involves three major processes namely, preprocessing, outlier detection, and classification. In the first step i.e., data preprocessing, there exists three stages namely, format conversion, class labelling, and missing value replacement using k-nearest neighbors (KNN) model. Outliers are superior values which considerably varies from other data observations. So, it might represent the variability in measurement, experimental errors or novelty too. So, Hierarchical Clustering (HC)-based outlier detection technique is incorporated in OD-DSAE model, and thereby classification performance can be improved. The proposed model was simulated using Python 3.6.5 on a dataset collected by the researcher themselves. A series of experiments was conducted and the results were investigated under different aspects. The experimental outcomes inferred that the OD-DSAE model has outperformed the compared methods and achieved high precision of 96.17%, recall of 98.69%, specificity of 89.50%, accuracy of 96.18%, and F-score of 97.41%.     

Effective K-nearest neighbor classifications for Wisconsin breast cancer data sets

ABSTRACT

Using machine learning algorithms for early prediction of the signs and symptoms of breast cancer is in demand nowadays. One of these algorithms is the K-nearest neighbor (KNN), which uses a technique for measuring the distance among data. The performance of KNN depends on the number of neighboring elements known as the K value. This study involves the exploration of KNN performance by using various distance functions and K values to find an effective KNN. Wisconsin breast cancer (WBC) and Wisconsin diagnostic breast cancer (WDBC) datasets from the UC Irvine machine learning repository were used as our main data sources. Experiments with each dataset were composed of three iterations. The first iteration of the experiment was without feature selection. The second one was the L1-norm based selection from the model, which used the linear support vector classifier feature selection, and the third iteration was with Chi-square-based feature selection. Numerous evaluation metrics like accuracy, receiver operating characteristic (ROC) curve with the area under curve (AUC) and sensitivity, etc., were used for the assessment of the implemented techniques. The results indicated that the technique involving the Chi-square-based feature selection achieved the highest accuracy with the Canberra or Manhattan distance functions for both datasets. The optimal K values for these distance functions ranged from 1 to 9. This study indicated that with the appropriate selection of the K value and a distance function in KNN, the Chi-square-based feature selection for the WBC datasets gives the highest accuracy rate as compared with the existing models.

Abbreviations: KNN: K-nearest neighbor; Chi2: Chi-square; WBC: Wisconsin breast cancer     

Piezoelectric properties of (K,Na)NbO3 thin films deposited on (001)SrRuO3/Pt/MgO substrates

(K(x),Na(1-x))NbO(3) (KNN) thin films were deposited on (001)SrRuO(3)/(001)Pt/(001)MgO substrates by RF-magnetron sputtering, and their piezoelectric properties were investigated. The x-ray diffraction measurements indicated that the KNN thin films were epitaxially grown with the c-axis orientation in the perovskite tetragonal system. The lattice constant of the c-axis increased with increasing concentrations of potassium. The KNN thin films showed typical ferroelectric behavior; the relative dielectric constant epsilon(r) was 270 to approximately 320. The piezoelectric properties were measured from the tip displacement of the KNN/MgO unimorph cantilevers; the transverse piezoelectric coefficient epsilon*(31) (= d(31)/s(E)(11)) of KNN (x = 0) thin films was calculated to be -0.9 C/m(2). On the other hand, doping of potassium caused an increase in the piezoelectric properties, and the KNN (x = 0.16) films showed a relatively large transverse piezoelectricity of epsilon*(31) = -2.4 C/m(2).     

不同烧结工艺对K0.5Na0.5NbO3陶瓷微结构的影响

采用普通烧结方法和热压烧结方法制备了K0.5Na0.5NbO3(KNN)无铅压电陶瓷.着重研究了两种烧结工艺对陶瓷的微观结构、晶粒形貌及致密度的影响.研究结果表明,两种烧结方法制备的陶瓷样品都具有单一的正交钙钛矿结构,与普通烧结工艺相比,利用热压烧结工艺制备的样品呈现较高的相对密度(大于98%)、较小的晶粒尺寸(0.6μm左右)及较低的介电损耗(1 kHz,tanδ≤2.8%).实验中发现对于热压烧结的样品,通过改变后期退火温度,样品的晶粒尺寸,致密度可以有规律地变化.     

Computer and Information Sciences-II

A simple non-iterative procedure is described for obtaining missing value estimates by solving a set of simultaneous linear equations that can be written directly. The method is derived for the two-way crossed classification and results for the P-way crossed classification are also given. Following the procedure of Yates (1933), estimates of the missing values are obtained such that their residuals are zero, thereby minimizing the error mean square. For the P-way crossed classifications the error mean square may be a pooled mean square of higher ordered interactions. The procedure is applicable to other designs as well.     

Computational intelligence techniques for human brain MRI classification

This study proposes an image classification methodology that automatically classifies human brain magnetic resonance (MR) images. The proposed methods contain four main stages: Data acquisition, preprocessing, feature extraction, feature reduction and classification, followed by evaluation. First stage starts by collecting MRI images from Harvard and our constructed Egyptian database. Second stage starts with noise reduction in MR images. Third stage obtains the features related to MRI images, using stationary wavelet transformation. In the fourth stage, the features of MR images have been reduced using principles of component analysis and kernel linear discriminator analysis (KLDA) to the more essential features. In last stage, the classification stage, two classifiers have been developed to classify subjects as normal or abnormal MRI human images. The first classifier is based on K‐Nearest Neighbor (KNN) on Euclidean distance. The second classifier is based on Levenberg‐Marquardt (LM‐ANN). Classification accuracy of 100% for KNN and LM‐ANN classifiers has been obtained. The result shows that the proposed methodologies are robust and effective compared with other recent works.     

锰掺杂和氧化铌种子层对铌酸钾钠薄膜电性能的影响

采用溶胶凝胶法制备铌酸钾钠(KNN)系薄膜,根据薄膜的微观形貌结构、电学性能和漏电机制研究了Mn掺杂和种子层对KNN薄膜性能的影响。结果表明：锰掺杂能显著提高薄膜的铁电性能和降低漏电流;在薄膜与衬底之间加入氧化铌种子层,使薄膜的电学性能进一步提高。薄膜的漏电流机制由空间电荷传导和欧姆传导转为欧姆传导和肖特基发射,使漏电流和剩余极化值减小。在强度为600 kV/cm的电场中,有种子层且掺10%(摩尔分数)Mn的KNN薄膜,其最大极化值和剩余极化值分别为20.33 µC/cm²和2.94 µC/cm²。     

基于机器学习的开放创新社区创意识别方法研究

目的 探索基于机器学习的开放创新创意识别方法,解决创意识别过程中存在的耗时长、效率低、成本高等问题。方法 从用户特征、用户参与度和创意内容特征三个方面构建评估模型,以OpenIDEO社区为研究对象,采集数据并进行数据清洗和数据转化映射,最后进行多种机器学习算法的参数优化,并以F1值为选择标准,选择分类效果最佳的算法作为分类模型。结果 运用KNN、SVM、决策树、随机森林四种机器学习算法分析OpenIDEO数据,随机森林算法通过参数优化取得了最大的F1值（0.919 09）,同时对于验证数据,该算法同样可以取得较好的分类效果。结论 应用机器学习方法对开放式创新社区中的创意进行识别,具有较高的可行性和有效性,可以大大降低社区在创意筛选中的投入,提高创新效率,优化社区生态。