基于Bagged CART和Boosted CART的高光谱影像分类技术研究

本文针对高光谱影像数据光谱分辨率高,数据量大的特点,采用以CART决策树为弱分类器的Bagging和Boosting集成学习算法对该影像进行分类,通过实验分析比较,体现出了Bagged CART和Boosted CART算法用于分类时的有效性和准确性。     

聚类支持下决策树模型在动车组空调故障检测中的应用

随着高铁的快速发展,人们对乘车环境的要求也越来越高。列车空调系统运行于列车高速振动环境下,性能极易下降且定期检修成本巨大,高效准确的空调故障检测尤为重要。本文用K-means聚类对空调数据进行分类,找到明显异常的类别并与业务部门确认后找到真正的空调系统故障数据,并对故障数据进行标记,利用CART决策树建立空调故障检测模型。     

阻燃包装纸研发技术优化研究

目的 为探索机器学习算法利用检验大数据快速鉴别复合包装膜袋材质的可行性。方法 以不同复合层数、不同功能层材质、不同食品接触层材质的10种复合包装膜袋共计1333个样本作为数据集,将韧性向拉伸强度、刚性向拉伸强度、韧性向断裂标称应变、刚性向断裂标称应变、水蒸气透过率、氧气透过率、厚度等7个维度的性能测试数据作为特征值,利用人工智能机器学习算法进行复合包装膜袋材质鉴别。结果 综合比较决策树、逻辑回归、支持向量机、K近邻、神经网络、高斯朴素贝叶斯等6种学习算法后,发现决策树算法的准确率和kappa系数最高,运行速度也很快。经参数优化后,决策树算法的鉴别结果准确率为95.4%,kappa系数为93.2%。结论 决策树算法在复合包装膜袋材质鉴别中具有一定优势。     

面向基因数据分类的旋转森林算法研究

针对基因表达数据高维和小样本的特点,介绍一种基于主成分分析的决策树集成分类算法——旋转森林.首先通过对数据属性集的随机分割,再对子集进行主成分分析变换,保留全部的主成分系数,重新组成一个稀疏矩阵.然后对变换后的数据利用非剪枝决策树集成算法进行分类.再结合ReliefF算法,选用3组基因表达数据验证算法,对比Bagging决策树和随机森林两种集成方法.结果表明旋转森林算法对基因数据具有更好的分类精度,同时验证旋转森林在较低的集成数的情况下,可以取得良好的效果.     

沿河县土地利用现状与效益分析

随着西部大开发的深入实施和“十二五”规划的新要求,我县的新型工业化及城镇化进程,社会主义新农村建设步伐将进一步加快,经济社会发展水平也会进一步提高。随着我县社会经济的不断发展,根据我县2005年-2010年以来的土地变更资料分析我县土地利用结构、土地利用特点和土地利用效益,提出合理利用土地的新思路。     

基于遥感的区域土地利用格局动态分析

以太原市1990年和2000年的Landsat TM影像为基本图件,在遥感和地理信息系统技术支持下,用土地利用指数定量分析了该区域土地利用的时空格局和演变趋势,并对其原因进行了初步探讨.结果表明:①耕地和林地是研究区域内居优势地位的土地类型,但这两种类型正不断地向其他土地类型流转,建设用地呈大幅度增长,水域面积变化较小.②10年来研究区域内的土地利用类型发生了复杂的相互转换,尤以林草交错带和农草交错带变动剧烈.③10年来区域内耕地、林地、草地、水域以及未利用土地的重要度均有所下降,城镇的重要度则有较大幅度的增加.城镇用地已经取代耕地而成为区域内重要度最高的土地利用类型.④10年来全区土地利用破碎度呈增加趋势,土地利用图斑数不断增加,但城镇的破碎度反而有所降低.⑤人口与经济的增长、城市化进程的加快是引起研究区土地利用格局变化的最主要因素.     

基于SEM-PCA-DT的多联机制冷剂泄漏在线故障诊断

制冷剂泄漏会给多联机带来不良影响,需要对其进行实时且准确的故障诊断。本文基于结构方程原理和试验数据,利用数据预处理提取的特征变量,建立结构方程模型,与决策树模型对比,分析结构方程在选择重要变量上的正确性,并利用重要显变量采用主元分析的方法建立专家变量,代入决策树模型,其在线诊断正确率达到96.96%,相对单独采用决策树模型提高2.81%。     

业界动态

<正>2015年中国软包装市场消费量达607万吨数据显示,全球消费市场对软包装市场需求将维持加速增长的趋势,其中发展中国家的需求占比将继续扩大。作为软包装消费量最大的国家,2015年中国的市场消费量达到了607万吨。亚洲是最大的消费区域,占全球总量的40%,2015~2020年,软包装消费量将以年均6.6%的速度增长。印度与中     

ID3决策树分类算法在学生信息中的应用

将数据挖掘中ID3决策树分类算法应用于高校教学环节中的大量数据信息,从中获取知识,继而对高校教学进行研究,可以起到科学地指导教学,提高教学质量的目的。     

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。     

Traffic analysis zone level crash estimation models based on land use characteristics

The objective of this paper is to develop crash estimation models at traffic analysis zone (TAZ) level as a function of land use characteristics. Crash data and land use data for the City of Charlotte, Mecklenburg County, North Carolina were used to illustrate the development of TAZ level crash estimation models. Negative binomial count models (with log-link) were developed as data was observed to be over-dispersed. Demographic/socio-economic characteristics such as population, the number of household units and employment, traffic indicators such as trip productions and attractions, and, on-network characteristics such as center-lane miles by speed limit were observed to be correlated to land use characteristics, and, hence were not considered in the development of TAZ level crash estimation models. Urban residential commercial, rural district and mixed use district land use variables were observed to be correlated to other land use variables and were also not considered in the development of the models. Results obtained indicate that land use characteristics such as mixed use development, urban residential, single-family residential, multi-family residential, business and, office district are strongly associated and play a statistically significant role in estimating TAZ level crashes. The coefficient for single-family residential area was observed to be negative, indicating a decrease in the number of crashes with an increase in single-family residential area. Models were also developed to estimate these crashes by severity (injury and property damage only crashes). The outcomes can be used in safety conscious planning, land use decisions, long range transportation plans, and, to proactively apply safety treatments in high risk TAZs.     

Relationship between land use composition and PM<Subscript>10</Subscript> concentrations in Iskandar Malaysia

Air pollution has always been a concern with increasing urbanization and poor land use planning adding to the problem. This study sets out to investigate the relationship between land use composition of an area and its ambient concentration of 10-μm-or-less particulate matter (PM₁₀). For this study, Iskandar Malaysia has been chosen as the study area. To compensate for the limited number of air quality monitoring stations in the study area, Terra MODIS aerosol optical depth Level 2 products are used to assess PM₁₀ concentration spatially. Land use data were developed from LANDSAT images used together with the land use database from the local authority. Finally, the relationship between land use composition and concentrations of PM₁₀ in the study area are explained using contour ternary plots. The plots show how different compositions of three major urban land uses (residential, commercial, industrial) in an area results in different concentration levels of PM₁₀. Concentrations of PM₁₀ are evidently more affected by commercial land use, followed by industrial land use. Hence, responsible authorities in Iskandar Malaysia could control or reduce air pollution in an area by planning a better land use composition.     

On the application of remote sensing towards the estimation of cultivated land lost to urbanization

In this research work, a 40-km² SPOT-5 High-Resolution Imagery (HRI) of the Warsak locality in district Peshawar, Pakistan, was utilized to approximate the quantity of cultivated land lost to urbanization, due to the construction of new homes and buildings. The imagery from a period of 2005 to 2015 for wheat crop was taken, specifically during the months of March and June when the crop is rich green and golden ripe respectively. eCognition ® program’s Object-Oriented Classification Method (OOCM) was employed for recognition of land versus buildings. Nearest Neighbour (NN), Support Vector Machine (SVM), Decision Trees (DT) and Random Forests (RF) were utilized for the classification process. The results demonstrated that the urbanized area had increased by approximately 28 per cent in the area considered. Moreover, the efficacy of the proposed method is depicted by an accuracy of 97.9 per cent and a Kappa Statistics of 0.975 for the SVM classifier.     

A trend analysis on Sri Lankan politics based on facebook user reactions

The rapid growth of user interactions in social media sites gives useful insights in many areas. Facebook is the most popular social media site lately, with the highest number of active users, which is a valuable and hassle-free source in obtaining data. Despite its enthusiastic nature, it is a mere fact that people use Facebook to gain instant updates on the current state of affairs. The ability of getting updates from several sources of news channels in a single user news feed, the extreme ease of providing feedbacks on those news posts using gesture-based reactions, send and share messages among people are some of the main reasons for its increasing popularity in the perspective of attaining news. Politics has always been a ubiquitous topic in the world. Sri Lanka was in a war on terrorism for nearly three decades, followed by a governance (2005–2015) led by the same political party which was alleged for autocracy and lasted for nearly a decade has influenced the citizens’ political conviction heavily. On such a background, the “Good-Governance” (2015–2019) which is a coalition government, trounced the ruling government at the presidential election held by then, which they claimed to direct Sri Lanka towards a sustainable, stable, responsible and moral society with necessary constitutional amendments guaranteeing democracy to all ethnic groups eradicating corruption, wastage and fraud. The interest and motivation of this study builds up to discover whether there are any significant trends in the Sri Lankan political context following this transformation, in the perspective of the general public. Facebook user reactions on news posts have been used for this study as the data source. The analysis of this study reveals an increasing trend of user reactions in politics from 2011 to 2018. Further, it is identified that the present government (2015–2019) has a decreasing trend of user reactions over the past years (2011–2015) in the sight of its citizens, although they pledged for a better governance. On the contrary, the previous government has an increasing trend even though they were overpowered by the “Good-Governance” for its alleged unscrupulous ruling.     

中国南方草地资源及其发展战略

中国南方草地区为秦岭、淮河以南、青藏高原以东的广大地区 ,包括西南岩溶山地灌丛草业生态经济区和东南常绿阔叶林一丘陵灌丛草业生态经济区 .大面积的连片草地 ,主要分布在西南岩溶地区 .南方草地大体可以分为云贵高原区、华东区和华南区 .中国南方草地的发展可以划分为三类系统耦合生态经济带 ,即西南岩溶山地系统耦合生态经济带、江淮平原丘陵系统耦合生态经济带和岭南沿海平原丘陵生态经济带 .每一个系统耦合经济带包含中心城市、农村生产基地和科学技术支持系统 .文中介绍的发展模式不仅有效地开发利用南方草山草坡和三荒地 ,而且成功地增加了农田的面积 ,这对于人多地少的中国南方来说 ,具有重要的意义     

Estimating windblown PM-10 emissions from vacant urban land using GIS

This paper presents a Geographic Information Systems (GIS)-based methodology to estimate annual area-wide airborne particulate matter with an aerodynamic diameter of less than 10 microm (PM-10) emissions, and identify zones with high emissions in order to efficiently implement mitigation strategies. Application of the methodology is demonstrated using the land disposal boundary within Clark County, NV as the study area, which is currently classified as a non-attainment area by United States Environmental Protection Agency (US EPA). The estimated PM-10 emissions depend on the extent of disturbed vacant land area, undisturbed vacant land area, emission factors by soil group, and wind speeds. Portable wind tunnel field test data were used to estimate emission factors at 78 sites in the study area. Portable wind tunnel results were categorized by the wind speed range and the corresponding site soil group in order to estimate emission factors by soil group and the wind speed range. Wind speed data were obtained from the Clark County Health District's air quality monitoring stations. The proximal area over which the wind speeds are same is obtained by constructing "Thiessen" polygons around each wind speed monitoring station. PM-10 emissions were estimated as a function of the extent of disturbed vacant lands, the measured or estimated erodibility of the soil surfaces, and the intensity, duration and frequency of erosive wind events.     

Factor complexity of crash occurrence: An empirical demonstration using boosted regression trees

Factor complexity is a characteristic of traffic crashes. This paper proposes a novel method, namely boosted regression trees (BRT), to investigate the complex and nonlinear relationships in high-variance traffic crash data. The Taiwanese 2004–2005 single-vehicle motorcycle crash data are used to demonstrate the utility of BRT. Traditional logistic regression and classification and regression tree (CART) models are also used to compare their estimation results and external validities. Both the in-sample cross-validation and out-of-sample validation results show that an increase in tree complexity provides improved, although declining, classification performance, indicating a limited factor complexity of single-vehicle motorcycle crashes. The effects of crucial variables including geographical, time, and sociodemographic factors explain some fatal crashes. Relatively unique fatal crashes are better approximated by interactive terms, especially combinations of behavioral factors. BRT models generally provide improved transferability than conventional logistic regression and CART models. This study also discusses the implications of the results for devising safety policies.     

中厚板加速冷却和直接淬火时冷却能力研究

为了确定加速冷却或直接淬火时实现预期的冷却速率所需的对流换热系数,利用MSC.MARC有限元分析软件对Q345B中厚钢板冷却过程中温度场进行了数值模拟计算.确定了实现直接淬火条件下不同厚度(20 mm)钢板的理论极限冷却速率所需的对流换热系数为15 000 W/(m2·℃),并分析了冷却速率与对流换热系数、钢板厚度之间关系.研究表明,对于同一厚度、材质中厚钢板,其冷却速率随对流换热系数的增加而增大.超快速冷却或直接淬火时,带钢冷却速率随对流换热系数增加而显著增加;对流换热系数大于15 000 W/(m2·℃)时,厚度(30 mm)钢板的冷却速率基本不变,达到其物理极限冷却速率;换热系数增加,厚度方向上温度梯度增加.     

中国生物质原料资源

对我国的有机废弃物(包括作物秸秆、林业剩余物、畜禽粪便、林业采伐及加工剩余物、育林剩余物、工业有机废弃物和城市有机垃圾)及可用于生物质原料植物生产的边际性土地(包括宜垦和宜林的荒地、现有的能源林地与农地)的相关资料进行了系统收集、整理与分析,分别提出了它们的资源量及可经济收集量。按2007年生物质产出现状与潜力,中国生物质资源的年产能为9.32亿tce,其中有机废弃物和边际性土地分别占41.1%和58.9%;预测2030年生物质原料资源的年产能为11.71亿tce,其中有机废弃物与边际性土地分别占48.3%和51.7%。