基于主成分与聚类分析法的城市土地利用规划分区研究——以广州市为例

[目的]土地利用分区是土地利用总体规划的核心内容之一,是合理确定土地利用方向,优化土地利用结构与布局的重要途径。为了对广州市土地利用及未来发展方向提供理论指导,结合其土地利用现状和特点,选取土地利用结构、土地开发利用程度、社会经济状况和环境资源承载力4项一级指标及15项二级指标,建立土地利用综合分区指标体系。[方法]通过SPSS19.0软件,运用主成分分析法提取出前4个主成分,并将其作为新的样本数据,采用层次聚类分析法对广州市土地利用进行综合分区。[结果]根据聚类分区结果,将广州市分为4类土地利用区:中心发达区,包括荔湾区、海珠区、天河区和越秀区;综合发展区,包括白云区、番禺区、花都区、增城区和从化区;重点潜力区即南沙区;资源组合区是(新)黄埔区。根据各分区的差异性,提出各区域土地资源开发利用中存在的主要问题及未来发展方向。[结论]根据各区的自然经济条件及土地利用状况,将广州市分为4类土地利用区具有合理性,能为广州市今后土地利用规划的科学编制提供参考。     

基于改进指纹聚类的WLAN定位优化方法

将K-means聚类算法应用到无线局域网（WLAN）位置指纹定位中，虽然可以缩短定位时间，但是容易降低定位精度。为了解决此问题，提出了基于改进指纹聚类的WLAN定位优化方法。首先根据接收信号强度标准差来优化初始聚类中心的选取,然后对指纹数据进行聚类处理,最后进行在线定位。实验结果表明，与传统的WLAN位置指纹定位方法和K-means聚类定位方法相比，基于改进指纹聚类的定位优化方法不仅缩短了定位时间，还能有效提高定位精度。     

基于PSO的信息熵数据融合非均匀分簇路由算法

针对无线传感器网络分簇算法中能量分布不均衡导致的"热区"和簇头负载过重问题,提出了一种基于PSO算法优化簇头选举的非均匀分簇算法。在候选簇头选举和竞争半径计算过程中综合考虑节点动态能量、节点密度和节点距基站距离,将网络进行非均匀分簇,并引入PSO算法进行最终簇头选举。根据节点能量、节点密度和距基站距离确定簇间单跳多跳结合的路由规则,选取代价函数小的节点作为下一跳节点。基于节点信息熵确定融合阈值,进行簇内数据融合剔除冗余数据。仿真结果表明,改进算法的数据传输量比EEUC算法和UCRA算法分别提高了20%和10%,提升了数据的融合效率,有效延长了网络生命周期,簇头能量消耗得到均衡,减少了网络能量消耗,网络的整体性能显著优于其他对比算法。     

基于因子与聚类分析的江苏省邮政业务协同运作实效研究

中国邮政业务出现了发展缓慢、停滞甚至倒退的现象,邮政业务创新成为不可回避的现实问题。通过2013年江苏邮政通信企业统计数据,对江苏13个邮政局的邮政各类业务协同发展水平状况以及资产配置水平进行因子分析和聚类分析。得出江苏省邮政业务地区发展的层次分布情况,分析了业务发展差距的原因,并总结了邮政在协同发展邮政业务的经验。     

一种基于自适应相似矩阵的谱聚类算法

为了消除在构建谱聚类算法的相似矩阵时,高斯核函数中尺度参数的波动影响,构建了一种自适应相似矩阵,并应用到谱聚类算法中。自适应相似矩阵中数据点间的距离度量采用测地距离算法,相距较近的两点间的距离近似于欧氏距离,相距较远的两点则先根据欧氏距离得到每个数据点的k个近邻点,然后累加近邻点的测地距离,由此得到每对数据点间的最短距离。两点间的局部密度用共享近邻的定义来表示,更好地刻画了数据集的本征结构。在5个人工数据集和国际通用UCI数据库中的5个真实数据集上进行实验。实验结果表明,所提算法的聚类准确率高于对比算法的准确率,对复杂分布数据有很强的自适应能力。研究成果为数据挖掘及机器学习提供了思路和方法。     

利用改进特征金字塔模型的SAR图像多目标船舶检测

深度学习模型中的特征金字塔网络（Feature Pyramid Network，FPN）常被用作合成孔径雷达（Synthetic Aperture Radar，SAR）图像中多目标船舶的检测。针对复杂场景下多目标船舶检测问题，提出了一种基于改进锚点框的FPN模型。首先将特征金字塔模型嵌入传统的RPN（Region Proposal Network）并映射成新的特征空间用于目标检测，然后利用基于形状相似度距离（Shape Similar Distance,SSD）度量的Kmeans聚类算法优化FPN的初始锚点框，并使用SAR船舶数据集测试。实验结果表明，所提算法目标检测精确率达到98.62%，在复杂场景下与YOLO、Faster RCNN、FPN based on VGG/ResNet等模型进行对比，模型准确率提高，整体性能更好。     

Macroeconomic Announcements and the Distribution of Price-Endings in the U.S. Treasury Market

We investigate how new information impacts quote clustering in the bond market. We find that clustering, along with quote activity, price volatility and bid-ask spreads, increases sharply in the minutes following releases of macroeconomic news. Each returns to near-normal levels within the hour. Effects are strongest for more liquid on-the-run notes and for the announcements typically associated with substantial information flow. The strong positive comovement of clustering, quote activity, price volatility, and bid-ask spreads supports the conclusion that innovations of these variables are endogenous to the arrival and incorporation of information into prices.     

Analysis of the Service Dominant Logic network,authors, and articles

The present study analyses research conducted in Service Dominant Logic (S-D logic), providing an overview of its intellectual structure. For that purpose, a structured approach was adopted through the implementation of bibliometric methods, using co-citation analysis combined with performance analysis and an integrative science mapping approach, which considered a timeline, a graph, and a distance-based perspective. This allowed the identification of the leading authors and of six clusters of articles based on bibliographic coupling, along with the most prominent research themes clustered in five subdomains of research based on co-occurrence of terms. The networks extracted show the associations between the main articles and concepts treated by the S-D logic community, unveiling the salient challenges and prospective expansion facing its future theoretical and practical context.     

Early-warning analysis for currency crises in emerging markets: A revisit with fuzzy clustering

Currency crises, also often called balance-of-payment crises, occur when massive capital outflows force a country to devalue or float its currency. The world-wide integration of capital markets since the 1980s and 1990s has increased the degree of capital mobility, which also determined a substantial turbulence in foreign exchange markets and frequent currency crises. In this paper, we explore advanced supporting instruments for predicting currency crises, based on an empirical study of the currency crisis episodes in 23 emerging markets around the world during the second half of last century. More specifically, we investigate the usefulness of prediction models built based on the fuzzy c-means method. First we build clustering models that partition data into a certain number of overlapping natural groups. Thereafter, we classify the data clusters into early-warning clusters and tranquil clusters. We compare the performance of our models with a conventional c-means clustering model and a benchmark probit model. The results show that the proposed models achieve a similar level of out-of-sample performance as the probit model and c-means model. The fuzzy approach also introduces additional explanatory advantages into the early-warning analysis process. Copyright © 2010 John Wiley & Sons, Ltd.     

集群企业创新行为机制分析

在传统产业中小企业集群中,普遍存在的企业模仿为什么没有抑制企业创新?结合相关文献分析和对现实中集群发展历程及企业行为的现察,可以发现其中的内生性模仿机制和竞争性、合作性创新机制,从企业行为机制视角观察集群创新是如何启动的;在企业不同阶段和不同集群类型中,企业模仿和创新的路径也不同.由此得出了培育产业集群和升级产业集群方面的政策含义,指出政府对于传统产业集群创新升级方面大有作为.