Support vector classifier based on principal component analysis

Support vector classifier （SVC） has the superior advantages for small sample learning problems with high dimensions, with especially better generalization ability. However there is some redundancy among the high dimensions of the original samples and the main features of the samples may be picked up first to improve the performance of SVC. A principal component analysis （PCA） is employed to reduce the feature dimensions of the original samples and the pre-selected main features efficiently, and an SVC is constructed in the selected feature space to improve the learning speed and identification rate of SVC. Furthermore, a heuristic genetic algorithm-based automatic model selection is proposed to determine the hyperparameters of SVC to evaluate the performance of the learning machines. Experiments performed on the Heart and Adult benchmark data sets demonstrate that the proposed PCA-based SVC not only reduces the test time drastically, but also improves the identify rates effectively.

PROJECTION PURSUIT PRINCIPAL COMPONENT ANALYSIS AND ITS APPLICATION TO METEOROLOGY

Projection Pursuit (PP) Principal Component Analysis (PCA) method is herein introduced and applied to the field of meteorology for the first time. Some problems relevant to meteorological application are dis- cussed in detail and comparisons with EOF method are made with the emphasis on robustness.

Face Recognition Based on Support Vector Machine and Nearest Neighbor Classifier

Support vector machine (SVM), as a novel approach in pattern recognition, has demonstrated a success in face detection and face recognition. In this paper, a face recognition approach based on the SVM classifier with the nearest neighbor classifier (NNC) is proposed. The principal component analysis (PCA) is used to reduce the dimension and extract features. Then one-against-all stratedy is used to train the SVM classifiers. At the testing stage, we propose an al-

Nitrogen application and intercropping change microbial community diversity and physicochemical characteristics in mulberry and alfalfa rhizosphere soil

Intercropping of mulberry(Morus alba L.)and alfalfa(Medicago sativa L.)is a new forestry-grass compound model in China,which can provide high forage yields with high protein.Nitrogen application is one of the important factors determining the production and quality of this system.To elucidate the advantages of intercropping and nitrogen application,we analyzed the changes of physicochemical properties,enzyme activities,and microbial communities in the rhizosphere soil.We used principal components analysis(PCA)and redundancy discriminators analysis to clarify the relationships among treatments and between treatments and environmental factors,respectively.The results showed that nitrogen application significantly increased pH value,available nitrogen content,soil water content(SWC),and urea(URE)activity in rhizosphere soil of monoculture mulberry.In contrast,intercropping and intercropping+N significantly decreased pH and SWC in mulberry treatments.Nitrogen,intercropping and intercropping+N sharply reduced soil organic matter content and SWC in alfalfa treatments.Nitrogen,intercropping,and intercropping+N increased the values of McIntosh diversity(U),Simpson diversity(D),and Shannon-Weaver diversity(H’)in mulberry treatments.However,PC A scatter plots showed clustering of monoculture mulberry with nitrogen(MNE)and intercropping mulberry without nitrogen(M0).Intercropping reduced both H’and D but nitrogen application showed no effect on diversity of microbial communities in alfalfa.There were obvious differences in using the six types of carbon sources between mulberry and alfalfa treatments.Nitrogen and intercropping increased the numbers of sole carbon substrate in mulberry treatments where the relative use rate exceeded 4%.While the numbers declined in alfalfa with nitrogen and intercropping.RDA indicated that URE was positive when intercropping mulberry was treated with nitrogen,but was negative in monoculture alfalfa treated with nitrogen.Soil pH and SWC were positive with mulberry treatments but were negative with alfalfa treatments.Intercropping with alfalfa benefited mulberry in the absence of nitrogen application.Intercropping with alfalfa and nitrogen application could improve the microbial community function and diversity in rhizosphere soil of mulberry.The microbial community in rhizosphere soil of mulberry and alfalfa is strategically complementary in terms of using carbon sources.

Face Recognition Combining Eigen Features with a Parzen Classifier

A face recognition scheme is proposed, wherein a face image is preprocessed by pixel averaging and energy normalizing to reduce data dimension and brightness variation effect, followed by the Fourier transform to estimate the spectrum of the preprocessed image. The principal component analysis is conducted on the spectra of a face image to obtain eigen features. Combining eigen features with a Parzen classifier, experiments are taken on the ORL face database.

A computer aided detection framework for mammographic images using fisher linear discriminant and nearest neighbor classifier

Today, mammography is the best method for early detection of breast cancer. Radiologists failed to detect evident cancerous signs in approximately 20% of false negative mammograms. False negatives have been identified as the inability of the radiologist to detect the abnormalities due to several reasons such as poor image quality, image noise, or eye fatigue. This paper presents a framework for a computer aided detection system that integrates Principal Component Analysis (PCA), Fisher Linear Discriminant (FLD), and Nearest Neighbor Classifier (KNN) algorithms for the detection of abnormalities in mammograms. Using normal and abnormal mammograms from the MIAS database, the integrated algorithm achieved 93.06% classification accuracy. Also in this paper, we present an analysis of the integrated algorithm’s parameters and suggest selection criteria.

The Establishment and Application of a Kraken Classifier for Salmonella Plasmid Sequence Prediction

Introduction:Salmonella is a key intestinal pathogen of foodborne disease,and the plasmids in Salmonella are related to many biological characteristics,including virulence and drug resistance.A large number of plasmid contigs have been sequenced in bacterial draft genomes,however,these are often difficult to distinguish from chromosomal contigs.Methods:In this study,three different customized Kraken databases were used to build three different Kraken classifiers.Complete genome benchmark datasets and simulated draft genome benchmark datasets were constructed.Five-fold cross-validation was used to evaluate the performance of the three different Kraken classifiers by two benchmark datasets.Results:The predictive performance of the classifier based on all National Center for Biotechnology Information plasmids and Salmonella complete genomes was optimal.This optimal Kraken classifier was performed with Salmonella isolated in China.The plasmid carrying rate of Salmonella in China is 91.01%,and it was found that the Kraken classifier could find more plasmid contigs and antibiotic resistance genes(ARGs)than results derived from a plasmid replicon-based method(PlasmidFinder).Moreover,it was found that in the strains carrying ARGs,plasmids carried more ARGs[three,95%confidence interval(CI):1–14]than chromosomes(one,95%CI:1–7).Discussion:We found building a high-quality customized database as a Kraken classifier to be ideal for the prediction of Salmonella plasmid sequences from bacterial draft genomes.In the future,the Kraken classifier established in this study will play a significant role in ARG monitoring.

DM-L Based Feature Extraction and Classifier Ensemble for Object Recognition

Deep Learning is a powerful technique that is widely applied to Image Recognition and Natural Language Processing tasks amongst many other tasks. In this work, we propose an efficient technique to utilize pre-trained Convolutional Neural Network (CNN) architectures to extract powerful features from images for object recognition purposes. We have built on the existing concept of extending the learning from pre-trained CNNs to new databases through activations by proposing to consider multiple deep layers. We have exploited the progressive learning that happens at the various intermediate layers of the CNNs to construct Deep Multi-Layer (DM-L) based Feature Extraction vectors to achieve excellent object recognition performance. Two popular pre-trained CNN architecture models i.e. the VGG_16 and VGG_19 have been used in this work to extract the feature sets from 3 deep fully connected multiple layers namely “fc6”, “fc7” and “fc8” from inside the models for object recognition purposes. Using the Principal Component Analysis (PCA) technique, the Dimensionality of the DM-L feature vectors has been reduced to form powerful feature vectors that have been fed to an external Classifier Ensemble for classification instead of the Softmax based classification layers of the two original pre-trained CNN models. The proposed DM-L technique has been applied to the Benchmark Caltech-101 object recognition database. Conventional wisdom may suggest that feature extractions based on the deepest layer i.e. “fc8” compared to “fc6” will result in the best recognition performance but our results have proved it otherwise for the two considered models. Our experiments have revealed that for the two models under consideration, the “fc6” based feature vectors have achieved the best recognition performance. State-of-the-Art recognition performances of 91.17% and 91.35% have been achieved by utilizing the “fc6” based feature vectors for the VGG_16 and VGG_19 models respectively. The recognition performance has been achieved by considering 30 sample images per class whereas the proposed system is capable of achieving improved performance by considering all sample images per class. Our research shows that for feature extraction based on CNNs, multiple layers should be considered and then the best layer can be selected that maximizes the recognition performance.

The Application of Principal Component Analysis to the Identification of Fagaceae Leaf Fossils

Abundant fossil records show that the Fagaceae has remained a dominant component in the Northern Hemisphere since the Cenozoic. However, due to the large number of living species, it is not easy to identify leaves to a particular species. Consequently, the identification of fossil leaves belonging to the Fagaceae is problematic.

On the Application of Classified English Teaching in Vocational School

Over the past years,with the increasing enrollment of high school,vocational schools are facing great challenge for their existence and development,concerning the low proficiency of the students and great gap among them.The traditional English teaching mode which employs the same teaching contents,same teaching methods and teaching aims cannot satisfy students with different English levels.Therefore,in order to change the present situation,this paper proposes a new English teaching mode:classified English teaching.In the new mode,different students will be taught by different materials,different methods and with different aims.It can stimulate students'enthusiasm in English learning,and make every student develop appropriately.

A STRESS VECTOR-BASED CONSTITUTIVE MODEL FOR COHESIONLESS SOIL (Ⅱ)-APPLICATION

The stress vector-based constitutive model for cohesionless soil, proposed by SHI Hong-yan et al., was applied to analyze the deformation behaviors of materials subjected to various stress paths. The result of analysis shows that the constitutive model can capture well the main deformation behavior of cohesionless soil, such as stress-strain nonlinearity, hardening property, dilatancy, stress path dependency, non-coaxiality between the principal stress and the principal strain increment directions, and the coupling of mean effective and deviatoric stress with deformation. In addition, the model can also take into account the rotation of principal stress axes and the influence of intermediate principal stress on deformation and strength of soil simultaneously. The excellent agreement between the predicted and measured behavior indicates the comprehensive applicability of the model.

教育内卷的反思与破解:教授校长十年的实践探索

笔者从2015年开始做中小学校长,创办过苏州大学实验学校、第二实验学校和高邮实验学校。在主持苏州第三中学慧成实验项目时,笔者尝试探索学生父母不内卷的路径;在苏州大学实验学校任校长时,全面探索教师和学生不内卷的办法;在主持苏州中学伟长实验项目时,深入探索学生不内卷的教学变革;如今在任苏州大学师范学院院长,又在积极推动“大国良师工作坊”,推动大学老师到中小学中去破除内卷。在基础教育园地里,笔者从一个思考者转变为一个实践者,从一名大学老师转变为一名基础教育管理者,一直在不断思考、探索中国基础教育内卷的破解之道。

食品检测中近红外光谱分析技术的应用研究

为了实现对食品成分、品质等方面的准确检测,为食品安全监管和质量控制提供有效的技术支持,引入近红外光谱分析技术,开展了该项技术在食品检测中的应用研究。通过样品制备、选择光谱采集设备、设置采集参数,采集食品近红外光谱,引入标准正态变量变换算法和自适应滤波算法,预处理采集的光谱数据,基于预处理后的数据,结合主成分分析法,构建食品成分特征与近红外光谱数据之间的数学模型,实现近红外光谱分析技术在食品检测的应用设计。实验结果表明,提出的研究应用后,食品成分检测结果与真实值更加接近,检测均方根误差较小,食品检测的准确性得到了显著提升。

不同性质园所园长任期结束综合督导省级评估指标设置研究

地方政府应通过了解管理区域内政策需求,积极推进制度创新,提高政策执行效能。分析发现,我国地方政府对园长以及幼儿园的督导指标具有较大的一致性。本文以行政管理自由裁量权理论为指导,基于福建省幼儿园办园行为督导评估数据分析,并借鉴其他相关政策文件内容,探讨不同性质园所园长任期结束综合督导评估指标优化设置问题。研究发现,办园行为督导数据分析可以为分类评估指标优化设置提供政策需求线索;公办园和普惠性民办园在园舍规划、设备设施、玩具图书与课程材料、专业成长、薪资待遇这5项指标方面差异显著,显示出分类评估指标优化设置的潜在需求,因此提出针对公、民办园园长评估分类指标的5项建议,并建言对已有评估资料进行深入分析,循证支持对不同性质园所的分类管理。

委托代理理论视角下科技查新的共生应用研究

文章旨在通过委托代理理论视角研究科技查新的共生应用,探讨委托代理关系下科技查新的可能性和限制,为企业和个人在查新过程中提供理论指导。同时,通过委托代理理论视角科技查新的共生应用,为企业和个人在查新过程中提供了新的思路和方法。

浙江省灵江流域地表水化学特征及灌溉适宜性

[目的]灵江是浙南诸河流域重要河流,在浙南一带具有典型性和代表性。分析灵江流域地表水体的水化学特征和演化机理及其灌溉适宜性,为该区域生态保护和高质量发展提供科学依据。[方法]综合利用数理统计、Piper三线图、主成分分析和离子比等方法对流域地表水体进行水化学统计分析及成因判别;通过绘制Wilcox图和USSL图评估流域内地表水体灌溉适宜性。[结果]①灵江流域地表水的化学类型在空间上具有分带性。从中上游(Ⅰ区)-下游温黄平原河网区(Ⅱ区)-台州湾入海口(Ⅲ区)水化学类型从HCO_(3)-Ca型向Cl-Na型过渡。②Ⅰ区地表水的化学特征主要受到岩石风化作用中的硅酸盐岩溶解影响,少量受碳酸盐岩溶解影响;Ⅱ区地表水的水化学特征主要受硅酸盐岩溶解影响,Ⅲ区地表水的化学特征主要受蒸发盐岩溶解影响。③Ⅰ,Ⅱ区地表水的K^(+),Na^(+)主要来源于硅酸盐矿物溶解,Ca^(2+),Mg^(2+),HCO^(-)_(3),SO^(2-)_(4)主要受硅酸盐岩溶解,少量来源于碳酸盐岩的溶解。NO^(-)_(3)则主要来源于人类活动。④Ⅰ,Ⅱ区地表水适用于农业灌溉,Ⅲ区地表水灌溉适宜性较差,易引起盐碱害。[结论]灵江流域地表水体的水化学组分受天然溶解及人类活动共同影响,中上游和下游温黄平原河网区地表水适宜农业灌溉,台州湾入海口地表水易引起盐碱害,在制定农业灌溉及生态保护和高质量发展规划时应予以重视。

含萘高性能聚酯关键单体专利分析

基于专利文献检索与分析,对国内含萘高性能聚酯关键单体的专利保护特点进行研究,分析了2,6-NDA/2,6-NDC技术的专利申请年趋势、法律状态、研究类型、申请人分布等,并重点对主要申请人的专利技术布局进行剖析。国内有关2,6-NDA/2,6-NDC技术的专利申请量在2000年大幅增长,以2,6-NDA的制备和纯化技术研究为主。中国石化在该领域的专利技术布局十分突出,自2012年至今持续在氧化合成、酯化和产物纯化方面开展研究。株式会社晓星在2,6-NDA制备工艺及设备、NDC纯化工艺上进行了少量专利布局,但有效专利占比高。煤炭科学技术研究院2022年的专利申请活跃度较高,专利布局主要围绕2,6-NDC纯化工艺和设备。

牛粪与化肥配施比例对苹果园土壤有机碳库和酶活性的影响

【目的】研究牛粪化肥配施比例对苹果园土壤活性有机碳组分和酶活性的影响,揭示不同配施比例对土壤碳库的影响,为苹果园牛粪与化肥的科学配施及土壤质量改善提供理论支撑。【方法】选取中国农业科学院果树研究所砬山施肥试验平台中不施肥(CK)、100%化肥(CF100)、25%牛粪配施75%化肥(CM25CF75)、50%牛粪配施50%化肥(CM50CF50)、75%牛粪配施25%化肥(CM75CF25)、100%牛粪(CM100)6个处理,分别测定施肥区(F)与非施肥区(NF)土壤活性有机碳组分(颗粒有机碳,POC;微生物量碳,MBC;易氧化有机碳,ROC)和酶活性(α-D-葡萄糖苷酶,AG;β-D-葡萄糖苷酶,BG;纤维素酶,CBH;过氧化物酶,PER;脲酶,UR)等相关指标。【结果】(1)牛粪化肥配施后施肥区土壤SOC、POC和ROC的含量随着牛粪施入比例的增加而增加;CM50CF50处理的MBC含量最高,比CK提高了139.7%。非施肥区,CF100、CM25CF75、CM50CF50、CM75CF25处理与CK相比土壤POC含量分别下降了32.8%、28.4%、21.6%、14.7%,CM50CF50和CM75CF25处理ROC含量分别下降了31.5%和17.4%。同一处理施肥区活性有机碳组分含量明显高于非施肥区。(2)CM25CF75、CM50CF50、CM75CF25、CM100处理施肥区土壤α-D-葡萄糖苷酶活性较CK分别提高了87.7%、68.4%、278.1%、331.6%;CM25CF75处理β-D-葡萄糖苷酶活性最高(39.00μg·g^(-1)·h^(-1));脲酶活性随牛粪施入比例的增加先升高后降低。牛粪化肥配施非施肥区土壤α-D-葡萄糖苷酶和脲酶活性也显著提高。(3)牛粪化肥配施显著提高了施肥区土壤POC/SOC和碳库管理指数(CPMI),CM25CF75、CM50CF50、CM75CF25、CM100处理的碳库管理指数分别提高了19.7%、38.3%、56.2%、103.5%。非施肥区牛粪化肥配施碳库管理指数显著下降。同一处理施肥区土壤碳库管理指数明显高于非施肥区。(4)相关分析和主成分分析表明,施肥区土壤中活性有机碳组分与α-D-葡萄糖苷酶活性呈显著正相关,牛粪比例的提高对土壤活性有机碳的提高贡献较大。【结论】牛粪化肥配施对于苹果园施肥区土壤的改善效果大于非施肥区。高比例的牛粪(配施比例在50%以上)能够提高土壤有机碳各组分含量和促进土壤酶活性增强,是苹果园科学施肥的较好的配施比例。

基于人工智能的专利技术分析

该文针对人工智能领域相关申请分别进行全球、中国的专利态势分析,包括申请量逐年变化趋势、主要申请人分析及技术构成等,建立全链的思维导图。通过分析国内外人工智能技术专利布局的内在原因,为我国关键核心技术创新及产业发展奠定理论基础。

基于主成分特征的相似微藻分类算法的研究

海洋微藻是一种光合自养型生物,随着水体富营养化日益严重,藻类大量繁殖逐渐形成水华和赤潮,如何快速准确识别有益藻和有害藻是当务之急。但是许多藻类形态彼此相近,难以分辨。为了解决这个问题,文中首先将输入的图像进行标准化处理,计算图像第一主成分载荷特征;其次设计一个基于逻辑回归的二分类模型,分两类标签进行训练;最后构造代价函数以及sigmoid函数,根据梯度下降法得出的预测结果与实际情况对比,计算最终准确率为92.86%。与广泛使用的藻类二分类算法SVM分类器比较结果显示,在相似藻的分类精度上提高了1.86%。