文本聚类算法的分析与比较

聚类是一种重要的文本信息处理方法,文章介绍了常用的文本聚类算法,从这些算法的适用范围、初始参数的影响、终止条件以及对噪声的敏感性等方面对它们进行了分析与比较.对文本聚类算法的应用有一定的指导意义.     

城市大气环境监测优化布点模糊优选模型及应用实例

文章根据城市大气环境监测优化布点的模糊性，提出了一种模糊环境条件下的模糊聚类与模糊识别理论模型，并且在山东省肥城等市成功应用。结合统计方法确定出的大气监测优化点位，不但具有代表性和整体性，而且具有较高的分辨率，能快速准确地、最大范围地反映出该区域的环境空气质量状况、大气污染扩散规律、污染源分布特点、污染气象及地理位置特征，为环境管理和政府决策提供可靠依据。     

用模糊混合聚类法对重金属潜在毒性排序

本文采用模糊混合聚类方法,即最大矩阵元原理与模糊ISODATA聚类分析相结合的方法,在计算机上对51个重金属元素的物理、化学性质数据进行了分析,并用筛选出的十项与毒性有关的数据对重金属潜在毒性进行了计算、分类和排序。     

城市环境质量多级模糊综合评价

文章将模糊聚类与层次分析相结合，提出了城市环境质量多级模糊综合评价法，并将该方法应用于宣州市环境质量综合评价，结果表明，该方法克服了综合指数法受人为因素影响大的缺点，较好地反映了环境质量分级界限的模糊性，并且较好地解决了权值分配问题，使评价结论更合理，可靠，是一种有价值的城市环境质量综合评价方法。     

南京市主要绿化树种对大气固体悬浮物净化能力及规律研究

对南京市不同功能绿地类型的26种主要绿化树种净化大气固体悬浮物的能力进行监测分析,研究不同功能区主要树种叶片滞尘量的季节变化,应用SAS软件,综合考虑单位面积滞尘量、单叶滞尘量、干重滞尘量3种指标对绿化树种滞尘能力进行聚类评估.结果表明,大部分绿化树种叶片滞尘量的季节变化为春季高、夏季降低、秋季增高、冬季达最高,这与空气中悬浮颗粒物含量季节变化的规律相一致;同一树种在不同地点的叶片滞尘量可以反映树木所在环境空气质量,灵谷寺公园和栖霞山空气质量优于梅山钢铁厂和宁镇公路;不同树木类型间滞尘能力差异较大,灌木滞尘量最大,常绿乔木其次,落叶乔木最小.     

长沙市典型园林植物叶片的滞尘等级与模式识别研究

基于长沙市10种典型植物叶片的多个滞尘指标,运用模糊数学理论构建植物的滞尘等级与识别模型,实现绿化植物滞尘综合量化评判方法的探究.首先,常规方法采集叶样后,运用打孔称质量法测量叶表面积参数,显微观察法计算植物叶表气孔与绒毛密度,干洗法测量单位叶面积最大滞尘量与自然滞尘量.然后,运用模糊聚类理论对10种植物滞尘指标进行滞...     

一种基于三维指标因子的流域多年径流丰枯k-means聚类法

径流丰枯聚类研究的传统方法多建立在年径流量的单一指标之上,容易导致分析的片面化。针对这个问题,提出了衡量流域多年径流丰枯状态的三维指标因子及权重,将其耦合到k-means聚类法的相似度计算与收敛分析中,在此基础上对对多年径流进行丰枯聚类,构建了一种基于三维指标因子的流域多年径流丰枯k-means聚类法。以该方法对广东省鉴江流域下游化州站1956—2006年的径流系列进行聚类分析,并与基于年径流量单一指标的k-means聚类方法进行对比,结果表明该方法是较全面且符合实际的。     

北亚热带过渡区土壤的模糊聚类分析

本文阐述了北亚热带过渡区十个土壤剖面的成土条件和基本性质，应用模糊聚类对土壤的五项诊断指标进行了分析，并与发生分类和诊断分类相比较．在中国土壤系统分类中提出建立淋溶土纲和始成土纲的基础上，探讨了供试土壤的分类地位。其中０１－０４号剖面属于典型铁质湿润淋溶上亚类；０５—０７和０９号剖面归为铁质粘磐半干润淋溶土亚类；０８号剖面为典型半干润淡色始成上亚类，１０号剖面为典型淡色滞水淋溶士亚类。     

河南省大气污染的气候分类

分析确定了影响大气污染的主要气候因子。根据省内各气象站的常规测资料，采用模糊聚类和纳污指数的方法将该省分成５个气候类型区。结果表明，纳污能力以中部地区为最强，占全省面积３／５的南、东、北邻近省界地区较强，豫西山区最弱。     

电解锰产业集聚区河流锰污染演变趋势和时空分布特征

松桃河流域是我国锰产业最为集中的区域之一,资源开发和利用形成的污染严重.通过对松桃河电解锰产业集聚区流域2015-2019年的8个典型断面的总锰监测数据的分析,用秩相关系数法探究了河流锰污染总体变化趋势,并用系统聚类分析法分析了河流锰的时空分布特征,结合聚类结果对污染趋势做进一步分析.秩相关系数分析结果表明,近年来松桃...     

Bayesian model-based cluster analysis for predicting macrofaunal communities

To predict macrofaunal community composition from environmental data a two-step approach is often followed: (1) the water samples are clustered into groups on the basis of the macrofauna data and (2) the groups are related to the environmental data, e.g. by discriminant analysis. For the cluster analysis in step 1 many hard, seemingly arbitrary choices have to be made that nevertheless influence the solution (similarity measure, clustering strategy, number of clusters). The stability of the solution is often of concern, e.g. in clustering by the program. In the discriminant analysis of step 2 it can occur that a water sample is misclassified on the basis of the environmental data but on further inspection happens to be a borderline case in the cluster analysis. One would then rather reclassify such a sample and iterate the two steps. Bayesian latent class analysis is a flexible, extendable model-based cluster analysis approach that recently has gained popularity in the statistical literature and that has the potential to address these problems. It allows the macrofauna and environmental data to be modelled and analyzed in a single integrated analysis. An exciting extension is to incorporate in the analysis prior information on the habitat preferences of the macrofauna taxa such as is available in lists of indicator values. The output of the analysis is not a hard assignment of water samples to clusters but a probabilistic (fuzzy) assignment. The number of clusters is determined on the basis of the Bayes factor. A standard feature of the Bayesian method is to make predictions and to assess their uncertainty. We applied this approach to a data set consisting of 70 water samples, 484 macrofauna taxa and four environmental variables for which previously a five cluster solution had been proposed. The standard for Bayesian estimation, the Gibbs sampler, worked fine on a subset with only 12 selected taxa but did not converge on the full set with 484 taxa. This is due to many configurations in which the assignment probabilities are all very close to either 0 or 1. This convergence problem is comparable with the local optima problem in classical cluster optimization algorithms, including the EM algorithm used in Latent Gold, a Windows program for latent class analysis. The convergence problem needs to be solved before the benefits of Bayesian latent class analysis can come to fruition in this application. We discuss possible solutions.     

福建省菜园土壤重金属的含量及其污染评价

采集了福建省26个县市158份菜园表层土壤样品,测定了土壤样品中重金属含量,采用单项及综合污染指数法对土壤重金属污染进行评价分析,并通过元素相关性分析和聚类分析,研究了菜园土壤重金属元素的赋存特征与区域分布.结果表明,福建省菜园土壤存在着不同程度的重金属污染,主要污染元素为Pb,Hg和Cd;元素相关性分析与聚类分析的结果表明:Cu,Zn,Ni和As具有伴随污染的特点,其污染源主要来自土壤母质和大量施用的农药和化肥;Cd,Pb和Hg在土壤中各聚成一类,说明它们在菜园土壤具有较独特的污染源,特别是Cd和Hg,其污染特征明显.而Pb和Cd虽有各自的污染源,但其污染具有一定的相关性.     

Multiple source spatial cluster detection via multi-criteria analysis

Multiple data sources are essential to provide reliable information regarding the emergence of potential health threats, compared to single source methods. Spatial Scan Statistics have been adapted to analyze multivariate data sources, but only ad hoc procedures have been devised to address the problem of selecting the most likely cluster and computing its significance. In this work, information from multiple data sources of disease surveillance is incorporated to achieve more coherent spatial cluster detection using tools from multi-criteria analysis. The best cluster solutions are found by maximizing two objective functions simultaneously, based on the concept of dominance. To evaluate the statistical significance of solutions, a statistical approach based on the concept of attainment function is used. The multi-criteria approach has several advantages: the representation of the evaluation function for each data source is clear, and does not suffer from an artificial, and possibly confusing mixture with the other data source evaluations; it is possible to attribute, in a rigorous way, the statistical significance of each candidate cluster; and it is possible to analyze and pick-up the best cluster solutions, as given naturally by the non-dominated set. The methodology is illustrated with real datasets.     

Combining the Species-Area-Habitat Relationship and Environmental Cluster Analysis to Set Conservation Priorities: a Study in the Zhoushan Archipelago, China

Abstract: Identification of priority areas is a fundamental goal in conservation biology. Because of a lack of detailed information about species distributions, conservation targets in the Zhoushan Archipelago (China) were established on the basis of a species–area–habitat relationship (choros model) combined with an environmental cluster analysis (ECA). An environmental‐distinctness index was introduced to rank areas in the dendrogram obtained with the ECA. To reduce the effects of spatial autocorrelation, the ECA was performed considering spatial constraints. To test the validity of the proposed index, a principal component analysis–based environmental diversity approach was also performed. The priority set of islands obtained from the spatially constrained cluster analysis coupled with the environmental‐distinctness index had high congruence with that from the traditional environmental‐diversity approach. Nevertheless, the environmental‐distinctness index offered the advantage of giving hotspot rankings that could be readily integrated with those obtained from the choros model. Although the Wilcoxon matched‐pairs test showed no significant difference among the rankings from constrained and unconstrained clustering process, as indicated by cophenetic correlation, spatially constrained cluster analysis performed better than the unconstrained cluster analysis, which suggests the importance of incorporating spatial autocorrelation into ECA. Overall, the integration of the choros model and the ECA showed that the islands Liuheng, Mayi, Zhoushan, Fodu, and Huaniao may be good candidates on which to focus future efforts to conserve regional biodiversity. The 4 types of priority areas, generated from the combination of the 2 approaches, were explained in descending order on the basis of their conservation importance: hotspots with distinct environmental conditions, hotspots with general environmental conditions, areas that are not hotspots with distinct environmental conditions, and areas that are not hotspots with general environmental conditions.     

Are ecological groups of species optimal for forest dynamics modelling?

The huge diversity of tree species in tropical rain-forests makes the modelling of its dynamics a difficult task. One-way to deal with it is to define species groups. A classical approach for building species groups consists in grouping species with nearby characteristics, using cluster analysis. A group of species is then characterized by the same list of attributes as a single species, and it is incorporated in the model of forest dynamics in the same way as a single species. In this paper, a new approach for building species group is proposed. It relies on the discrepancy between model predictions when all species are considered separately, and model predictions when species groups are used. An aggregation error that quantifies the bias in model predictions that results from species grouping is thus defined. We then define the optimal species grouping as the one that minimizes the aggregation error. Using data from a tropical rain-forest in French Guiana and a toy model of forest dynamics, this new method for species grouping is confronted to the classical method based on cluster analysis of the species characteristics, and to a combined method based on a cluster analysis that uses the aggregation error as a dissimilarity between species. The optimal species grouping is quite different from the classical species grouping. The ecological interpretation of the optimal groups is difficult, as there is no direct linkage between the species characteristics and the way that they are grouped. The combined approach yields species groups that are closed to the optimal ones, with much less computations. The optimal species groups are thus specific to the model of forest dynamics and lack the generality of those of the classical method, that in turn are not optimal.     

应用模糊数学评价和预测海河的水质状况

本文应用模糊数学的方法评价１９９１－１９９６年的海河水质状况，并建立高锰酸盐指数的灰色预测模型。     

Characteristics of clustered particles in skimming flows on a stepped spillway

Air–water flows at hydraulic structures are commonly observed and called white waters. The free-surface aeration is characterised by some intense exchanges of air and water leading to complex air–water structures including some clustering. The number and properties of clusters may provide some measure of the level of particle-turbulence and particle–particle interactions in the high-velocity air–water flows. Herein a re-analysis of air–water clusters was applied to a highly aerated free-surface flow data set (Chanson and Carosi, Exp Fluids 42:385–401, 2007). A two-dimensional cluster analysis was introduced combining a longitudinal clustering criterion based on near-wake effect and a side-by-side particle detection method. The results highlighted a significant number of clustered particles in the high-velocity free-surface flows. The number of bubble/droplet clusters per second and the percentage of clustered particles were significantly larger using the two-dimensional cluster analysis than those derived from earlier longitudinal detection techniques only. A number of large cluster structures were further detected. The results illustrated the complex interactions between entrained air and turbulent structures in skimming flow on a stepped spillway, and the cluster detection method may apply to other highly aerated free-surface flows.     

Can best competitors avoid extinction as habitat destruction?

A nonlinear n-population metapopulation model, which can describe the nonlinear relationship between one species and other species or between one species and the habitat, is presented in this paper. By simulation and mathematics analysis, we discover that species possess an ability to control or avoid extinction during habitat destruction. Any species in n-population metapopulation can increase (decrease) the influence of habitat destruction if it agrees (disagrees) with the environment, and it also can increase (decrease) the proportion of sites occupied by all species by harmonizing (not harmonizing) with the other species.     

Testing spatial cluster occurrence in maps equipped with environmentally defined structures

We propose a novel tool for testing hypotheses concerning the adequacy of environmentally defined factors for local clustering of diseases, through the comparative evaluation of the significance of the most likely clusters detected under maps whose neighborhood structures were modified according to those factors. A multi-objective genetic algorithm scan statistic is employed for finding spatial clusters in a map divided in a finite number of regions, whose adjacency is defined by a graph structure. This cluster finder maximizes two objectives, the spatial scan statistic and the regularity of cluster shape. Instead of specifying locations for the possible clusters a priori, as is currently done for cluster finders based on focused algorithms, we alter the usual adjacency induced by the common geographical boundary between regions. In our approach, the connectivity between regions is reinforced or weakened, according to certain environmental features of interest associated with the map. We build various plausible scenarios, each time modifying the adjacency structure on specific geographic areas in the map, and run the multi-objective genetic algorithm for selecting the best cluster solutions for each one of the selected scenarios. The statistical significances of the most likely clusters are estimated through Monte Carlo simulations. The clusters with the lowest estimated p-values, along with their corresponding maps of enhanced environmental features, are displayed for comparative analysis. Therefore the probability of cluster detection is increased or decreased, according to changes made in the adjacency graph structure, related to the selection of environmental features. The eventual identification of the specific environmental conditions which induce the most significant clusters enables the practitioner to accept or reject different hypotheses concerning the relevance of geographical factors. Numerical simulation studies and an application for malaria clusters in Brazil are presented.     

Space-time Bayesian small area disease risk models: development and evaluation with a focus on cluster detection

This paper extends the spatial local-likelihood model and the spatial mixture model to the space-time (ST) domain. For comparison, a standard random effect space-time (SREST) model is examined to allow evaluation of each model’s ability in relation to cluster detection. To pursue this evaluation, we use the ST counterparts of spatial cluster detection diagnostics. The proposed criteria are based on posterior estimates (e.g., misclassification rate) and some are based on post-hoc analysis of posterior samples (e.g., exceedance probability). In addition, we examine more conventional model fit criteria including mean square error (MSE). We illustrate the methodology with a real ST dataset, Georgia throat cancer mortality data for the years 1994–2005, and a simulated dataset where different levels and shapes of clusters are embedded. Overall, it is found that conventional SREST models fair well in ST cluster detection and in goodness-of-fit, while for extreme risk detection the local likelihood ST model does best.