基于SVM的综合实力评估系统研究

简略介绍了支持向量机的基本思想 ,研究其在综合评价中的应用。分析了对大学进行实力评估的过程 ,建立基于SVM的评价系统。比较新评价系统和采用其他方法如PCA (PrincipalComponentAnalysis主元分析 ) ,Fisher等建立的评价系统所分别取得的拟合效果。结果表明 :采用支持向量机设计的评价系统思路清晰 ,操作简单并且能取得更为理想的评估结果。     

一类非线性混合整规划的无约束连续化解法

将整系数多项式 0 - 1混合整规划问题化成无约束多项式规划问题。通过解该问题 ,能得到原问题的近似解。处理方法的特点是能够直接处理不等式约束情形 ,而不需要先将不等式约束化成等式约束再来处理     

银行个人住房抵押贷款预警系统构想

在分析银行住房抵押贷款的各种风险因素的基础之上，提出模糊预警模型，给出预警结果。     

略论艺术的分类法

本文论述了中外艺术的多元分类法,对其进行科学、系统地对照比较,并结合中外艺术发展的现状和艺术教育的实际情况提出自己的见解,即艺术四大门类分类法是当今较为科学、较易接受、较能应用的艺术分类法.并在四大门类划分的基础上,对各门类中的艺术科目进行了重新整合和构建.     

论为师育人之道

本文从理论与实践的结合上 ,阐述了教师教书育人的神圣职责 ,从“晓之以理、动之以情、教之以严、导之以行、授之以渔、示之以范”六个方面 ,探讨了师者育人之道。     

后新儒学及"公民儒学"相关问题之探讨

内圣之学应是在整个历史社会总体与生活世界中生长出来的学问,因此,面对整个外王情境的变化.内.圣之学必然应有所调整;而圣贤的教言也必须置于整个历史文化的总体情境下才能够有恰当的理解.面对心性之学与专制主义结合而造成的专制性、摹虐性及以理杀人等种种之异化,我们必须将原来儒学的内圣工夫转化为一套客观的制度结构,只有在契约性的社会结构之下,个人才能够自然地进入到社会展开公共对话.     

国内企业境外上市的动机及市场选择策略研究

文章分析了国内企业境外上市的现状、理论动机和现实动机,综合比较了境外主要资本市场特点,指出国内企业境外上市的资本市场选择因素主要包括境外交易所的定位、上市门槛及费用、市场规模和市场流动性.     

论“和谐”对仲裁制度价值目标的建构

建立和谐社会的价值目标,既解读继承历史,又立足创新社会,和谐的内在要求导向着仲裁制度的完善,和谐的价值目标奠定了仲裁的发展方向,是仲裁制度为之实现的终极目标;仲裁制度的固有属性也暗合了和谐的建构目的,并成为和谐社会目标下处理纷争的重要制度。     

信息时代行政法规的传播体制改革

20世纪 90年代 ,美、英等西方国家提出并建立的公共行政信息网络 ,有效地促进了政府行政立法、执法及社会公众进行行政监督 ,成为一种新的社会发展趋势。为此 ,借签西方国家的有益经验 ,提出加强我国公共行政信息网络建设的对策     

Tree health mapping with multispectral remote sensing data at UC Davis, California

Tree health is a critical parameter for evaluating urban ecosystem health and sustainability. Traditionally, this parameter has been derived from field surveys. We used multispectral remote sensing data and GIS techniques to determine tree health at the University of California, Davis. The study area (363 ha) contained 8,962 trees of 215 species. Tree health conditions were mapped for each physiognomic type at two scales: pixel and whole tree. At the pixel scale, each tree pixel within the tree crown was classified as either healthy or unhealthy based on vegetation index values. At the whole tree scale, raster based statistical analysis was used to calculate tree health index which is the ratio of healthy pixels to entire tree pixels within the tree crown. The tree was classified as healthy if the index was greater than 70%. Accuracy was checked against a random sample of 1,186 trees. At the whole tree level, 86% of campus trees were classified as healthy with 88% mapping accuracy. At the pixel level, 86% of the campus tree cover was classified as healthy. This tree health evaluation approach allows managers to identify the location of unhealthy trees for further diagnosis and treatment. It can be used to track the spread of disease and monitor seasonal or annual changes in tree health. Also, it provides tree health information that is fundamental to modeling and analysis of the environmental, social, and economic services produced by urban forests.