Spatial Data Mining Research Articles

A Comparative Study of Different Density based Spatial Clustering Algorithms

is an important descriptive model in data mining. It groups the data objects into meaningful classes or clusters such that the objects are similar to one another within the same cluster and are dissimilar to other clusters. Spatial clustering is one of the significant techniques in spatial data mining, to discover patterns from large spatial databases. In recent years, several basic and advanced algorithms have been developed for clustering spatial datasets. Clustering technique can be categorized into six types namely partitioning, hierarchical, density, grid, model, and constraint based models. Among these, the density based technique is best suitable for spatial clustering. It characteristically consider clusters as dense regions of objects in the data space that are separated by regions of low density (indicating noise).The clusters which are formed based on the density are easy to understand, filter out noise and discover clusters of arbitrary shape. This paper presents a comparative study of different density based spatial clustering algorithms, and the merits and limitations of the algorithms are also evaluated.

Deformation information system for facilitating studies of mining-ground deformations, development, and applications

Abstract. The paper presents the concept of the deformation information system (DIS) to support and facilitate studies of mining-ground deformations. The proposed modular structure of the system includes data collection and data visualisation components, as well as spatial data mining, modelling and classification modules. In addition, the system integrates interactive three-dimensional models of the mines and local geology. The system is used to calculate various parameters characterising ground deformation in space and time, i.e. vertical and horizontal displacement fields, deformation parameters (tilt, curvature, and horizontal strain) and input spatial variables for spatial data classifications. The core of the system in the form of an integrated spatial and attributive database has been described. The development stages and the functionality of the particular components have been presented and example analyses utilising the spatial data mining and modelling functions have been shown. These include, among other things, continuous vertical and horizontal displacement field interpolations, calculation of parameters characterising mining-ground deformations, mining-ground category classifications, data extraction procedures and data preparation preprocessing procedures for analyses in external applications. The DIS has been developed for the Walbrzych coal mines area in SW Poland where long-time mining activity ended at the end of the 20th century and surface monitoring is necessary to study the present-day condition of the former mining grounds.

Large-Scale Overlays and Trends: Visually Mining, Panning and Zooming the Observable Universe.

We introduce a web-based computing infrastructure to assist the visual integration, mining and interactive navigation of large-scale astronomy observations. Following an analysis of the application domain, we design a client-server architecture to fetch distributed image data and to partition local data into a spatial index structure that allows prefix-matching of spatial objects. In conjunction with hardware-accelerated pixel-based overlays and an online cross-registration pipeline, this approach allows the fetching, displaying, panning and zooming of gigabit panoramas of the sky in real time. To further facilitate the integration and mining of spatial and non-spatial data, we introduce interactive trend images-compact visual representations for identifying outlier objects and for studying trends within large collections of spatial objects of a given class. In a demonstration, images from three sky surveys (SDSS, FIRST and simulated LSST results) are cross-registered and integrated as overlays, allowing cross-spectrum analysis of astronomy observations. Trend images are interactively generated from catalog data and used to visually mine astronomy observations of similar type. The front-end of the infrastructure uses the web technologies WebGL and HTML5 to enable cross-platform, web-based functionality. Our approach attains interactive rendering framerates; its power and flexibility enables it to serve the needs of the astronomy community. Evaluation on three case studies, as well as feedback from domain experts emphasize the benefits of this visual approach to the observational astronomy field; and its potential benefits to large scale geospatial visualization in general.

Mining Complex Spatial Patterns: Issues and Techniques

Spatial data mining is the quantitative study of phenomena that are located in space. This paper investigates methods of mining patterns of a complex spatial data set (which generally describes any kind of data where the location in space of object holds importance). We based this research on the analysis of some spatial characteristics of certain objects. We began with describing the spatial pattern of events or objects with respect to their attributes; we looked at how to describe the spatial nature/characteristics of entities in an environment with respect to their spatial and non-spatial attributes. We also looked at modelling (predictive modelling/knowledge management of complex spatial systems), querying and implementing a complex spatial database (using data structure and algorithms). Critically speaking, the presence of spatial auto-correlation and the fact that continuous data types are always present in spatial data makes it important to create methods, tools and algorithms to mine spatial patterns in a complex spatial data set. This work is particularly useful to researchers in the field of data mining as it contributes a whole lot of knowledge to different application areas of data mining especially spatial data mining. It can also be useful in teaching and likewise for other study purposes.

Rural Architectural Intensification: A Multidisciplinar Planning Tool

When approaching a composite territorial problem that involves different scales and disciplines, it is necessary to establish a precise logical framework. Every planning or design activity is an iterative process applied to a complex system; not linear relations among the entities that compose the system are numerous and it is problematic to spell out them. Authors developed a framework that has a hybrid structure in which different classical tool such as Spatial Decision Support Systems, Knowledge Discovery and Data Mining (KDD), and Expert Systems (ES) converge. The method is not completely automatic and there is a continuous interaction between user and system. The main aim of the entire research group who participated to a national research (PRIN 2009) was to define an informed methodology for decision makers, stakeholders and public bureaus who have to (or want to) face the problem of improving and intensifying insediative activities in minor centers located in rural-urban context. In particular authors defined Rural Architectural Intensification (RAI) as a way to improve territorial features throughout a serie of interventions in small settlements and buildings. The explanation of the relations among different disciplines, different scales and different related methodologies is the key point of the paper. After an introduction and the description of RAI, authors introduce the main methodological structure; then each passage is detailed and specified considering the elements involved and the technical operations.

Physical Infrastructure Assessment for Emergency Medical Response

The performance of an emergency medical service (EMS) depends on the existing infrastructure and allocation of medical resources. It is challenging to implement an efficient EMS owing to the spatial complexity of the population and geographical layout within urban areas. Management of ambulance headquarters and hospitals, which are part of the physical infrastructure, should be assessed to ensure a timely response. The objective of this research is to propose a general assessment process for a prehospital EMS by evaluating the physical infrastructure. A geographic information system is adapted to visualize and analyze the EMS of a region EMS from various perspectives. In the assessment process, the EMS is evaluated with incident data and the transportation network is evaluated through spatial and temporal analyses such as service area, clustering, and spatial interpolation. As a case study, the process is applied to New Taipei City for the assessment of its prehospital EMS. The results have shown that the proposed assessment process can help discover existing deficiencies and offer suggestions to the current EMS. Regions less accessible by the EMS were discovered; the spatial coverage of historical demand incidents was 93.8% for the union of ambulance headquarters and 70.52% for the union of hospitals in their respective 3 km service areas. Cluster analysis suggested an EMS performance gain of 0.68% compared with the current situation for the construction of an additional ambulance facility and 1.45% for an additional hospital. Future work will extend the proposed process with specialized spatial and temporal data mining techniques and to apply the process to more case studies.

English

Spatial data mining or Knowledge discovery in spatial database is the extraction of implicit knowledge, spatial relations and spatial patterns that are not explicitly stored in databases. Co-location patterns discovery is the process of finding the subsets of features that are frequently located together in the same geographic area. In this paper, we discuss the different approaches like Rule based approach, Join-less approach, Partial Join approach and Constraint neighborhood based approach for finding co-location patterns.

Social-spatial structure of Beijing: a spatial-temporal analysis

In the past couple of post-reform decades of rapid urban growth of Chinese cities, Beijing experienced dramatic spatial and socio-economic transformation. This study aims to examine the evolution of Beijing’s social-spatial structure in the transforming process over the past 20 years. We first collect and process Beijing’s socio-demographic variables at the level of sub-district in three years between 1990 and 2009. Then a two-step spatial clustering approach is taken for spatial data mining of the spatial structures of social areas in the three time periods. Results are visualised in GIS. Finally, the interpretation of the analysis result is provided, followed by a discussion of policy implications for the development of a socially sustainable municipality.

A method for extracting rules from spatial data based on rough fuzzy sets

With the development of data mining and soft computing techniques, it becomes possible to automatically mine knowledge from spatial data. Spatial rule extraction from spatial data with uncertainty is an important issue in spatial data mining. Rough set theory is an effective tool for rule extraction from data with roughness. In our previous studies, Rough set method has been successfully used in the analysis of social and environmental causes of neural tube birth defects. However, both roughness and fuzziness may co-exist in spatial data because of the complexity of the object and the subjective limitation of human knowledge. The situation of fuzzy decisions, which is often encountered in spatial data, is beyond the capability of classical rough set theory. This paper presents a model based on rough fuzzy sets to extract spatial fuzzy decision rules from spatial data that simultaneously have two types of uncertainties, roughness and fuzziness. Fuzzy entropy and fuzzy cross entropy are used to measure accuracies of the fuzzy decisions on unseen objects using the rules extracted. An example of neural tube birth defects is given in this paper. The identification result from rough fuzzy sets based model was compared with those from two classical rule extraction methods and three commonly used fuzzy set based rule extraction models. The comparison results support that the rule extraction model established is effective in dealing with spatial data which have roughness and fuzziness simultaneously.

Application of the spatial data mining module in analysis of mining ground deformation factors

Spatial data mining methods for example those based on artificial neural networks (ANN) allow extraction of information from databases and detection of otherwise hidden patterns occurring in these data and in consequence acquiring new knowledge on the analysed phenomena or processes. One of these techniques is the multivariate statistical analysis, which facilitates identification of patterns otherwise difficult to observe. In the paper an attempt of applying self-organising maps (SOM) to explore and analyse spatial data related to studies of ground subsidence associated with underground mining has been described. The study has been carried out on a selected part of a former underground coal mining area in SW Poland with the aim to analyse the influence of particular ground deformation factors on the observed subsidence and the relationships between these factors. The research concerned the uppermost coal panels and the following factors: mining system, time of mining activity and inclination, thickness and depth below the ground of the exploited coal panels. It has been found that the exploratory spatial data analysis can be used to identify relationships in multidimensional data related to mining induced ground subsidence. The proposed approach may be found useful in identification of areas threatened by mining related subsidence and in creating scenarios of developing deformation zones and therefore aid spatial development of mining grounds.

Hybrid spatial data mining methods for site selection of emergency response centers

Traditional methods for site selection fail to give reasonable weight to the potential risk grade of different locations. Consequently, site selection does not highlight locations with higher risk grades. Thus, a hybrid analytical method for the site selection of emergency response centers is proposed. First, spatial predicates are incorporated into emergency event analysis. These spatial predicates describe the spatial relationships between emergency locations and surrounding objects. A spatial data association mining method is then developed to identify the correlation rules that contain emergency information and geographical factors. Such rules act as the weight adding mechanism on different urban buildings, such that every spatial object is assigned a risk grade based on these rules. Furthermore, a simulated annealing algorithm is developed by incorporating the weight adding method to locate the optimal sites for emergency centers. A series of experiments is conducted, and the results demonstrate the efficiency of the proposed hybrid method.

The Distribution Network Reconfiguration Based on Spatial Data Mining

Distribution network reconfiguration is one of the essential functions of the DMS system; it can be attributed to a number of constraints of large-scale nonlinear combinatorial optimization problem in mathematics. The characteristic of load change of time and space will affect the results of load forecasting. Distribution network reconfiguration relies on the load forecasting results. This paper proposes the more realistic distribution reconfiguration scheme based on the GIS system with space information through spatial data mining. For the real-time and efficiency of effective assurance data, the interface design in GIS system and distribution automation system are also proposed. The geographical information and real-time information are connected seamlessly, so that this two system information is highly unified. The solution can provide the data basis for distribution reconfiguration scheme accurately, and improve power supply reliability of distribution network. It shows that through the example: GIS based on the spatial data mining can provide load in quantity, time, space prediction for the deeper research of distribution network reconfiguration.

Study on Fuzzy Clustering Algorithm of Spatial Data Mining

With rapid development of space information crawl technology, different types of spatial database and data size of spatial database increases continuously. How to extract valuable information from complicated spatial data has become an urgent issue. Spatial data mining provides a new thought for solving the problem. The paper introduces fuzzy clustering into spatial data clustering field, studies the method that fuzzy set theory is applied to spatial data mining, proposes spatial clustering algorithm based on fuzzy similar matrix, fuzzy similarity clustering algorithm. The algorithm not only can solve the disadvantage that fuzzy clustering cant process large data set, but also can give similarity measurement between objects.

Analyzing the Wind Power Industry Competitiveness Based on SDM-TGSMN Method in Hebei Province

As a kind of new green renewable energy, wind power is being paid more attention to by the countries in the world. Wind energy resource is rich in Hebei Province, and wind energy has great potential, so it is important for energy sustainable development, harmonious development of social economy and ecological environment to develop and utilize wind power in Hebei Province. In order to analyze the wind power industry competitiveness in Hebei Province scientifically and accurately, this paper introduced the tree-structured growing self-organizing map network (TGSMN) into the spatial data mining (SDM) to be used in spatial clustering. This method not only can make up the limitation of the more data limit and biggish computation in the common spatial data mining, but overcome the restriction of traditional maps that must appoint in advance. The wind power industry competitiveness evaluation in Hebei Province showed that the results given by this method are reliable.

Degradation of cultivated bench terraces in the Three Gorges Area: Field mapping and data mining

Due to resettlements, construction of new infrastructure, and new land reclamation the rapid agricultural changes in the Three Georges Area (TGA) in Central China are expected to force the degradation of the cultivated terraced landscape. Consequently, increased soil erosion can hamper a sustainable land management in the mountainous TGA. This paper presents the model framework TerraCE (Terrace Condition Erosion) for determining the causes for different terrace conditions and terrace degradation based on field surveys and spatial data mining. For a total of 987 bench terrace plots in the Xiangxi catchment we collected data on their state of maintenance and terrace design to account for terrace stability and thus capability of soil conservation. Assessing the driving factors of terrace degradation was done by embedding terrain-based predictors and distance-transforms of remote-sensing data as indicators of environmental and anthropogenic influences. Random forests classification and regression models were applied for data mining. Terrace degradation in the Xiangxi catchment is obvious. The sequence of degradation ranges from ‘well maintained’ (21%), ‘fairly maintained’ (44%), and ‘partially collapsed’ (23%) to ‘completely collapsed’ (11%) terraces. The cross-validation error of the supervised TerraCE model is below 8%, allowing for reasonable and valid interpretations of the causes of terrace degradation. Data mining reveals indicators for anthropogenic effects such as the distance to settlements or to roads as major drivers for the spatial distribution of terrace conditions. The effect of relief, which can be regarded as the major natural driver for terrace degradation by erosive action is tributary but altered and overlaid by land use dynamics associated with the Three Gorges Dam. An important indicator representing a combined effect of terrain and human activity is the distance to stream channels of different orders. Applying TerraCE we reveal mechanisms of terrace degradation in disturbed environments and present a framework for standardized mapping and analysis of terrace degradation under cultivation. The approach might also be used to develop guidelines for terrace planning in mountainous terraced landscapes of limited carrying capacity, with respect to socio-economic as well as environmental conditions.

공간 데이터마이닝 분석을 통한 데이터의 효과적인 활용

Data mining is a useful technology that can support new discoveries based on the pattern analysis and a variety of linkages between data, and currently is utilized in various fields such as finance, marketing, medical. In this paper, we propose an effective utilization method of data based on analysis of spatial data mining. We make use of basic data of foreigners living in Seoul. However, the data has some features distinguished from other areas of data, classification as sensitive information and legal problem such as personal information protection. So, we use the basic statistical data that does not contain personal information. The main features and contributions of the proposed method are as follows. First, we can use Big Data as information through a variety of ways and can classify and cluster Big Data through refinement. Second. we can use these kinds of information for decision-making of future and new patterns. In the performance evaluation, we will use visual approach through graph of themes. The results of performance evaluation show that the analysis using data mining technology can support new discoveries of patterns and results.

GeoDMA—Geographic Data Mining Analyst

Remote sensing images obtained by remote sensing are a key source of data for studying large-scale geographic areas. From 2013 onwards, a new generation of land remote sensing satellites from USA, China, Brazil, India and Europe will produce in 1year as much data as 5 years of the Landsat-7 satellite. Thus, the research community needs new ways to analyze large data sets of remote sensing imagery. To address this need, this paper describes a toolbox for combing land remote sensing image analysis with data mining techniques. Data mining methods are being extensively used for statistical analysis, but up to now have had limited use in remote sensing image interpretation due to the lack of appropriate tools. The toolbox described in this paper is the Geographic Data Mining Analyst (GeoDMA). It has algorithms for segmentation, feature extraction, feature selection, classification, landscape metrics and multi-temporal methods for change detection and analysis. GeoDMA uses decision-tree strategies adapted for spatial data mining. It connects remotely sensed imagery with other geographic data types using access to local or remote database. GeoDMA has methods to assess the accuracy of simulation models, as well as tools for spatio-temporal analysis, including a visualization of time-series that helps users to find patterns in cyclic events. The software includes a new approach for analyzing spatio-temporal data based on polar coordinates transformation. This method creates a set of descriptive features that improves the classification accuracy of multi-temporal image databases. GeoDMA is tightly integrated with TerraView GIS, so its users have access to all traditional GIS features. To demonstrate GeoDMA, we show two case studies on land use and land cover change.

Change detection from remotely sensed images: From pixel-based to object-based approaches

The appetite for up-to-date information about earth’s surface is ever increasing, as such information provides a base for a large number of applications, including local, regional and global resources monitoring, land-cover and land-use change monitoring, and environmental studies. The data from remote sensing satellites provide opportunities to acquire information about land at varying resolutions and has been widely used for change detection studies. A large number of change detection methodologies and techniques, utilizing remotely sensed data, have been developed, and newer techniques are still emerging. This paper begins with a discussion of the traditionally pixel-based and (mostly) statistics-oriented change detection techniques which focus mainly on the spectral values and mostly ignore the spatial context. This is succeeded by a review of object-based change detection techniques. Finally there is a brief discussion of spatial data mining techniques in image processing and change detection from remote sensing data. The merits and issues of different techniques are compared. The importance of the exponential increase in the image data volume and multiple sensors and associated challenges on the development of change detection techniques are highlighted. With the wide use of very-high-resolution (VHR) remotely sensed images, object-based methods and data mining techniques may have more potential in change detection.

Clustering Algorithm for Spatial Data Mining: An Overview

data mining practice for the extraction of useful information and knowledge from massive and complex spatial database. Most research in this area has focused on efficient clustering algorithm for spatial database to analyze the complexity. This paper introduces an active spatial data mining approach that extends the current spatial data mining algorithms to efficiently support user-defined triggers on dynamically evolving spatial data. It shows that spatial data mining is a promising field, with fruitful research results and many challenging issues.

Understanding the spatial distribution of crime based on its related variables using geospatial discriminative patterns

Crime tends to cluster geographically. This has led to the wide usage of hotspot analysis to identify and visualize crime. Accurately identified crime hotspots can greatly benefit the public by creating accurate threat visualizations, more efficiently allocating police resources, and predicting crime. Yet existing mapping methods usually identify hotspots without considering the underlying correlates of crime. In this study, we introduce a spatial data mining framework to study crime hotspots through their related variables. We use Geospatial Discriminative Patterns (GDPatterns) to capture the significant difference between two classes (hotspots and normal areas) in a geo-spatial dataset. Utilizing GDPatterns, we develop a novel model—Hotspot Optimization Tool (HOT)—to improve the identification of crime hotspots. Finally, based on a similarity measure, we group GDPattern clusters and visualize the distribution and characteristics of crime related variables. We evaluate our approach using a real world dataset collected from a northeast city in the United States.