Spatial data mining system for ore-forming prediction

The authors designed the spatial data mining system for ore-forming prediction based on the theory and methods of data mining as well as the technique of spatial database,in combination with the characteristics of geological information data.The system consists of data management,data mining and knowledge discovery,knowledge representation.It can syncretize multi-source geosciences data effectively,such as geology,geochemistry,geophysics,RS.The system digitized geological information data as data layer files which consist of the two numerical values,to store these files in the system database.According to the combination of the characters of geological information,metallogenic prognosis was realized,as an example from some area in Heilongjiang Province.The prospect area of hydrothermal copper deposit was determined.

Spatial Multidimensional Association Rules Mining in Forest Fire Data

Hotspots (active fires) indicate spatial distribution of fires. A study on determining influence factors for hotspot occurrence is essential so that fire events can be predicted based on characteristics of a certain area. This study discovers the possible influence factors on the occurrence of fire events using the association rule algorithm namely Apriori in the study area of Rokan Hilir Riau Province Indonesia. The Apriori algorithm was applied on a forest fire dataset which containeddata on physical environment (land cover, river, road and city center), socio-economic (income source, population, and number of school), weather (precipitation, wind speed, and screen temperature), and peatlands. The experiment results revealed 324 multidimensional association rules indicating relationships between hotspots occurrence and other factors.The association among hotspots occurrence with other geographical objects was discovered for the minimum support of 10% and the minimum confidence of 80%. The results show that strong relations between hotspots occurrence and influence factors are found for the support about 12.42%, the confidence of 1, and the lift of 2.26. These factors are precipitation greater than or equal to 3 mm/day, wind speed in [1m/s, 2m/s), non peatland area, screen temperature in [297K, 298K), the number of school in 1 km2 less than or equal to 0.1, and the distance of each hotspot to the nearest road less than or equal to 2.5 km.

Spatial Data Mining to Support Environmental Management and Decision Making--A Case Study in Brazil

The growth of geo-technologies and the development of methods for spatial data collection have resulted in large spatial data repositories that require techniques for spatial information extraction, in order to transform raw data into useful previously unknown information. However, due to the high complexity of spatial data mining, the need for spatial relationship comprehension and its characteristics, efforts have been directed towards improving algorithms in order to provide an increase of performance and quality of results. Likewise, several issues have been addressed to spatial data mining, including environmental management, which is the focus of this paper. The main original contribution of this work is the demonstration of spatial data mining using a novel algorithm with a multi-relational approach that was applied to a database related to water resource from a certain region of S^o Paulo State, Brazil, and the discussion about obtained results. Some characteristics involving the location of water resources and the profile of who is administering the water exploration were discovered and discussed.

Scaling up the DBSCAN Algorithm for Clustering Large Spatial Databases Based on Sampling Technique

Clustering, in data mining, is a useful technique for discovering interesting data distributions and patterns in the underlying data, and has many application fields, such as statistical data analysis, pattern recognition, image processing, and etc. We combine sampling technique with DBSCAN algorithm to cluster large spatial databases, and two sampling based DBSCAN (SDBSCAN) algorithms are developed. One algorithm introduces sampling technique inside DBSCAN, and the other uses sampling procedure outside DBSCAN. Experimental results demonstrate that our algorithms are effective and efficient in clustering large scale spatial databases.

Mining multilevel spatial association rules with cloud models

The traditional generalization-based knowledge discovery method is introduced. A new kind of multilevel spatial association of the rules mining method based on the cloud model is presented. The cloud model integrates the vague and random use of linguistic terms in a unified way. With these models, spatial and nonspatial attribute values are well generalized at multiple levels, allowing discovery of strong spatial association rules. Combining the cloud model based method with Apriori algorithms for mining association rules from a spatial database shows benefits in being effective and flexible.

A New Method Based on Association Rules Mining and Geo-filter for Mining Spatial Association Knowledge

Association rule mining methods, as a set of important data mining tools, could be used for mining spatial association rules of spatial data. However, applications of these methods are limited for mining results containing large number of redundant rules. In this paper, a new method named Geo-Filtered Association Rules Mining(GFARM) is proposed to effectively eliminate the redundant rules. An application of GFARM is performed as a case study in which association rules are discovered between building land distribution and potential driving factors in Wuhan, China from 1995 to 2015. Ten sets of regular sampling grids with different sizes are used for detecting the influence of multi-scales on GFARM. Results show that the proposed method can filter 50%–70% of redundant rules. GFARM is also successful in discovering spatial association pattern between building land distribution and driving factors.

GIS-based spatial and temporal changes of land occupation caused by mining activities-a study in eastern part of Hubei Province

By using multi-source and multi-temporal high resolution remote sensing data and related techniques of remote sensing and geographic information systems, this paper analyzes the spatial and temporal changes of land occupation caused by mine development in four mining areas of eastern Hubei Province from 2011 to 2014, including Chengchao-Tieshan iron-copper polymetallic deposit area, Daye-Yangxin iron-copper polymetallic deposit area, E-Nan mining area, and Wuxue-Yangxin non-metallic mining area along the Yangtze River. The results show that: In the research area, land occupation of energy mine exploitation is small and in scattered distribution, with coal mine occupying the largest area, showing a downward trend in four years; land occupation of metal mines is large and in centralized distribution, with iron mine and copper mine occupying the largest area, showing a downward trend in four years; non-metallic mines are large and in great quantity, with mines of limestone for building and limestone occupying the largest area, showing a upward trend in four years.

Geospatial Area Embedding Based on the Movement Purpose Hypothesis Using Large-Scale Mobility Data from Smart Card

With the deployment of modern infrastructure for public transportation, several studies have analyzed movement patterns of people using smart card data and have characterized different areas. In this paper, we propose the “movement purpose hypothesis” that each movement occurs from two causes: where the person is and what the person wants to do at a given moment. We formulate this hypothesis to a synthesis model in which two network graphs generate a movement network graph. Then we develop two novel-embedding models to assess the hypothesis, and demonstrate that the models obtain a vector representation of a geospatial area using movement patterns of people from large-scale smart card data. We conducted an experiment using smart card data for a large network of railroads in the Kansai region of Japan. We obtained a vector representation of each railroad station and each purpose using the developed embedding models. Results show that network embedding methods are suitable for a large-scale movement of data, and the developed models perform better than existing embedding methods in the task of multi-label classification for train stations on the purpose of use data set. Our proposed models can contribute to the prediction of people flows by discovering underlying representations of geospatial areas from mobility data.

Research on spatial association rules mining in two-direction

In data mining from transaction DB, the relationships between the attributes have been focused, but the relationships between the tuples have not been taken into account. In spatial database, there are relationships between the attributes and the tuples, and most of the associations occur between the tuples, such as adjacent, intersection, overlap and other topological relationships. So the tasks of spatial data association rules mining include mining the relationships between attributes of spatial objects, which are called as vertical direction DM, and the relationships between the tuples, which are called as horizontal direction DM. This paper analyzes the storage models of spatial data, uses for reference the technologies of data mining in transaction DB, defines the spatial data association rule, including vertical direction association rule, horizontal direction association rule and two-direction association rule, discusses the measurement of spatial association rule interestingness, and puts forward the work flows of spatial association rule data mining. During two-direction spatial association rules mining, an algorithm is proposed to get non-spatial itemsets. By virtue of spatial analysis, the spatial relations were transferred into non-spatial associations and the non-spatial itemsets were gotten. Based on the non-spatial itemsets, the Apriori algorithm or other algorithms could be used to get the frequent itemsets and then the spatial association rules come into being. Using spatial DB, the spatial association rules were gotten to validate the algorithm, and the test results show that this algorithm is efficient and can mine the interesting spatial rules.

城市休闲产业聚类模式APM算法模型开发与校验

城市休闲相关产业的高质量发展对当前我国城市消费升级以及人居环境质量提升具有重要现实意义。但是,现有研究未能精准地捕捉海量广域分布的城市休闲产业的基本空间分布规律与结构,而已有的空间聚类算法较多适用于城市用地分析,未能很好地适用于离散分布的城市休闲产业研究。为此,文章基于空间兴趣点数据,开发距离通达值及空间集群中心点等算法,构建城市休闲旅游产业聚类模式空间算法模型(APM)。在以广州为例的研究中,APM模型捕捉出3170个以500 m步行生活圈为范围的城市休闲产业集群,校验了APM模型的科学性与应用价值。整体上,APM算法可以较好地捕捉城市休闲业态集群的空间结构,清晰识别城市休闲产业空间冷、热点分布的基本结构,由其捕捉行程的聚类边界与实际道路和建筑走向、水系边界、区域范围等重合度高,聚类集群符合实际情况,具备可信度与有效性。该研究是休闲产业集聚机制研究的一次方法创新,在算法精度、实际应用、可视化效率上均做出了创新性推进。与Fishnet方法相比,可以更科学精准地识别城市内部多个休闲消费商圈的边界,实现了高效率的城市休闲产业集群捕捉;与同位模型相比,可以呈现多类别的城市休闲业态结构,突破了现有研究只能捕捉两类业态组团的局限。

基于Voronoi图的空间点事件统计聚类方法

挖掘地理空间数据中点事件聚集模式对于揭示流行疾病、犯罪分布热点区域及城市基础设施空间分布格局等具有重要意义。针对不同形状、密度和大小的显著空间点聚集模式的识别,目前以空间扫描统计为代表的方法虽然可以对空间点聚类的显著性进行统计推断,减少虚假聚类结果,但其主要用于识别球形或椭圆形状的聚簇,对于沿着街道或河道分布的任意形状、不同密度的显著空间点聚簇识别还存在局限。因此,本研究提出一种基于Voronoi图的空间点聚集模式统计挖掘方法。首先,采用Voronoi图来度量空间点分布的聚集性,将空间点聚类问题转化为热点区域探测问题;其次,结合局部Gi*统计量探测统计上显著的空间点聚簇;最后,通过模拟数据和真实犯罪事件数据进行实验与对比分析。结果表明:本方法能够有效探测任意形状的空间点聚类,并对空间点簇的显著性进行统计判别,识别显著的空间点簇,减少随机噪声点的干扰;聚类识别结果优于现有代表性方法,如DBSCAN算法、空间扫描统计方法等。

基于网格空间团的多级同位模式挖掘方法

针对传统的多级同位模式挖掘方法未考虑到实际数据分布的网格特性,且从全局到区域的多级模式挖掘框架会导致算法效率低下的问题,提出逆向挖掘多级同位模式的新框架.先挖掘区域同位模式,再由区域同位模式推导出全局同位模式,提出有效的剪枝策略提高挖掘效率.考虑真实数据集中数据分布的网格特性,定义实例间的网格邻近关系,提出网格空间团及计算网格空间团的新颖方法.在区域划分阶段,提出基于自适应网格密度峰值聚类的区域划分方法,基于2阶网格空间团的网格相似性来分配簇.在合成和实际数据集上进行大量的实验,验证了提出方法的有效性、高效性和可扩展性,在真实数据集上的剪枝率可以达到78%.

SDML:基于空间数据库的空间数据挖掘语言

设计了一种基于空间数据库的空间数据挖掘语言SDML。根据SDML操作的对象以及挖掘过程的不同阶段 ,SDML语言可以分为视图操纵语言和模型操纵语言 ,分别负责对于数据挖掘视图和模型的操作。详细阐述了SDML的设计思想及其设计方案 ,针对空间泛化和空间关联这两个典型的空间数据挖掘问题 。

GIS与可视化SDM技术集成问题探讨

随着地理信息获取技术飞速发展 ,使得当前存储在空间数据库中的空间数据的深度和广度得到了前所未有的发展 .为了解决GIS目前面临的“数据爆炸但知识贫乏”的难题 ,在介绍GIS发展现状等相关问题、空间数据挖掘技术、可视化技术的基础上 ,分析了GIS中数据挖掘的过程、特点及其相关技术支持 ,探讨了可视化技术在GIS数据挖掘中的重要作用 .对GIS与可视化交互空间数据挖掘集成技术进行了初步的研究 ,分析阐述了GIS与SDM集成的必要性、集成模式和集成路线 ,提出了一个以GIS为中心的二者集成的体系结构 .

基于GIS与SDM技术的可视化空间数据分类研究

提出将GIS与可视化空间数据挖掘技术之集成的基本框架。在此基础上,基于VisualC++6.0和Ma-pObject2.0组件技术设计和开发了一个可视化交互空间数据挖掘分类系统,系统采用决策树方法和贝叶斯网络作为数据挖掘方法的基本算法,采用训练与学习相结合实现空间数据的分类。文中用实例数据对系统性能、算法和规则有效性进行了验证。结果表明,该系统是一个适用的、可扩展的可视化交互空间数据挖掘工具,系统能够实现数据挖掘实时动态的交互控制,实现了数据挖掘过程的可视化、挖掘模型的可视化和结果的可视化显示、可视化思考、可视化分析与评价。

基于改进列计算的空间并置模式挖掘方法

空间并置(co-location)模式挖掘旨在发现空间特征间的关联关系,是空间数据挖掘的重要研究方向。基于列计算的空间并置模式挖掘方法(CPM-Col算法)避开挖掘过程中最耗时的表实例生成操作,直接搜索模式的参与实例,成为当前高效的方法之一。然而,回溯法搜索参与实例仍是该方法的瓶颈,尤其在稠密数据和长模式下。为加速参与实例的搜索,充分利用CPM-Col算法搜索参与实例时得到的行实例,在不增加额外计算的前提下对CPM-Col算法进行两点改进。首先,将CPM-Col算法搜索到的行实例存储为部分表实例,利用子模式的部分表实例快速确定参与实例,避免了大量实例的回溯计算。其次,在CPM-Col算法获得一条行实例后,利用行实例的子团反作用于第一个特征,得到第一个特征的参与实例,避免了这些实例的回溯搜索。由此,提出了基于改进列计算的空间并置模式挖掘算法(CPM-iCol算法),并讨论了算法的复杂度、正确性和完备性。在合成数据和真实数据集上进行了实验,与经典的传统算法join-less和CPM-Col进行对比,CPM-iCol算法明显缩短了挖掘的时间,减少了回溯的次数。实验结果表明,该算法比CPM-Col具有更好的性能和可扩展性,特别在稠密数据集中效果更加明显。

基于遥感技术的矿山煤火燃烧风险评价

为了利用遥感技术研究矿山煤火燃烧风险与燃烧特征,以新疆五宫煤矿为研究区,利用国产高分6#卫星影像对五宫煤矿地面塌陷、煤岩裂隙等地质灾害进行解译,结合无人机热红外数据进行煤火燃烧评价因子的选取,采用层次分析法对评价因子进行权重赋值,使用空间分析、阈值分割方法进行煤火燃烧风险等级划分,结合实地调查,对火区成因、特征进行分析。结果表明:矿区存在5处煤火燃烧高风险区,其走向与含煤地层、井下开采走向一致。煤矿煤火燃烧特征有3种:露天开采引起的煤层燃烧,地质条件复杂,治理难度大;煤矸石堆热量积聚引起的煤火燃烧治理较容易,建议清除煤矸石后,做必要的防护措施;治理后煤层复燃引起的煤火燃烧,由于复燃原因复杂,建议进一步勘查再进行灭火治理。通过对矿山煤火风险性与燃烧类型进行研究,可以为矿山灭火治理和生态修复提供服务。

GIS与SDM集成构建土地定级专家信息系统的研究

提出将GIS与SDM技术集成应用于土地定级信息系统建设,介绍了GIS与SDM集成的基本框架。设计开发了一个基于GIS和SDM技术的土地定级专家信息系统,系统采用决策树方法作为数据挖掘方法的基本算法,通过决策树训练与学习相结合挖掘土地定级规则,运用专家系统推力技术进行匹配推理实现土地定级工作。结合武汉市商业用地类型土地定级实例数据对系统性能进行了验证。结果表明,系统具有良好的移植性、复用性、扩展性和广泛适应性的特点,运用此技术能较好地解决土地定级这种具有半结构和非结构化特点的问题。

基于多尺度时空图卷积网络的交通出行需求预测

满足公众高质量出行需求是智能交通系统(ITS)的主要挑战之一。目前,针对公共交通出行需求预测问题,现有模型大多采用固定结构的图描述出行需求的空间相关性,忽略了出行需求在不同尺度下具有不同的空间依赖关系。针对上述问题,提出一种多尺度时空图卷积网络(MSTGCN)模型。该模型首先从全局尺度和局部尺度构建全局需求相似图和局部需求相似图,这2种图可以捕获公共交通出行需求长期内较为稳定的全局特征和短期内动态变化的局部特征。利用图卷积网络(GCN)提取2种图中的全局空间信息和局部空间信息,并引入注意力机制融合两种空间信息。为了拟合时间序列中潜藏的时间依赖关系,利用门控循环单元(GRU)捕捉公共交通需求的时变特征。采用纽约市出租车订单数据集和自行车订单数据集进行实验,结果表明MSTGCN模型在自行车订单数据集上均方根误差(RMSE)、平均绝对误差(MAE)和皮尔逊相关系数(PCC)达2.7886、1.7371、0.7992,在出租车订单数据集上RMSE、MAE、PCC达9.5734、5.8612、0.9631。可见,MSTGCN模型可以有效地挖掘公共交通出行需求的多尺度时空特性,对未来公共交通出行需求进行准确预测。