Scheduled System Maintenance on May 29th, 2015:
IEEE Xplore will be upgraded between 11:00 AM and 10:00 PM EDT. During this time there may be intermittent impact on performance. We apologize for any inconvenience.
By Topic

Spatial Data Mining and Geographical Knowledge Services (ICSDM), 2011 IEEE International Conference on

Date June 29 2011-July 1 2011

Filter Results

Displaying Results 1 - 25 of 127
  • Positive spatial autocorrelation, mixture distributions, and geospatial data histograms

    Publication Year: 2011 , Page(s): 1 - 6
    Cited by:  Papers (4)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (228 KB) |  | HTML iconHTML  

    Researchers commonly construct histograms as a first step in representing and visualizing their geospatial data. Because of the presence of spatial autocorrelation in these data, these graphs usually fail to closely align with any of the several hundred existing ideal frequency distributions. The purpose of this paper is to address how positive spatial autocorrelation - the most frequently encountered in practice - can distort histograms constructed with geospatial data. Following the auto-normal parameter specification employed in WinBUGS for Bayesian analysis, this paper summarizes results for normal, Poisson, and binomial random variables (RVs) - three of the most commonly employed ones by geospatial scientists - in terms of mixture distributions. A spatial filter description of positive spatial autocorrelation is shown to approximate a normal distribution in its initial form, a gamma distribution when exponentiated, and a beta distribution when embedded in a logistic equation. In turn, these conceptualizations allow: the mean for a normal distribution to be distributed as a normal random variable (RV) with a zero mean and a specific variance; the mean for a Poisson distribution to be distributed as a gamma RV with specific parameters (i.e., a negative binomial distribution); and, the probability for a binomial distribution to be distributed as a beta RV with specific parameters (i.e., a beta-binomial distribution). Results allow impacts of positive spatial autocorrelation on histograms to be better understood. A methodology is outlined for recovering the underlying unautocorrelated frequency distributions. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Exploring non-stationarity of local mechanism of crime events with spatial-temporal weighted regression

    Publication Year: 2011 , Page(s): 7 - 12
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (485 KB) |  | HTML iconHTML  

    For a more effective understanding of dynamic of mechanism and cluster of local crime, this study uses kernel density to reveal abilities of detecting space-time hotspots in the context of time geography. Since spatial data are correlated in nature, geographically weighted regression (GWR) has been proven as an effective tool to address the spatial non-stationarity. Thus, this study adopts temporal variants to detect the spatial-temporal non-stationarity of structural measures simultaneously. Using a geocoded criminal dataset of residential burglary in Da-an District of Taipei City from 1999 to 2007, we examine the proposed framework allowing interactively 3-D visualization of crime hotspots by volume rendering. We also reveal the non-stationarity of estimations of social structural measures by a variant weighted regression approach. Emphasizing the supplementary aspect of our embedded framework, we conclude that 3-D spatial-temporal data analysis and the variant of geographically weighted regression could identify the space-time hotspots as well as extract and interpret the spatial-temporal non-stationarity of mechanism of residential burglary. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A similarity measure for time series of spatial lines intersection relations

    Publication Year: 2011 , Page(s): 13 - 15
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (115 KB) |  | HTML iconHTML  

    Spatial lines are common in various spatial applications such as GIS (Geographical Information System). Topological relationship is one of important properties between spatial lines. Neglecting disjoint relations and adjacent relations between spatial lines, in this paper we focus on intersection relations. Time series of spatial lines intersection relations record each spatial line intersection relations at each time. The changes of intersection relationship are caused by deformations (translation, rotation, and length of spatial lines) and increasing or decreasing of the number of spatial lines neighbours. We propose a similarity measure for time series of spatial lines topological relations. By this measure, we can cluster these spatial lines on time series of their intersection relations. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Study and analysis of temporal data using Hidden Markov models

    Publication Year: 2011 , Page(s): 16 - 20
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (381 KB) |  | HTML iconHTML  

    In this paper we have developed a method for dividing a set of temporal data into clusters by using Hidden Markov Models. Given a number of clusters, each cluster is represented by one Hidden Markov Model. In order to determine the optimal number of clusters and the consistency of their structures, this approach defines an objective function based on the calculation of likelihood. The algorithm is presented in terms of four nested levels of searches: (1) the search for the optimal number of clusters in a partition, (2) the search for the optimal structure for a given partition, (3) the search for the optimal HMM structure for each cluster, and (4) the search for the optimal HMM parameters for each HMM. Preliminary results are given to support the proposed methodology. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Using geographically weighted regression to explore the spatially varying relationship between land subsidence and groundwater level variations: A case study in the Choshuichi alluvial fan, Taiwan

    Publication Year: 2011 , Page(s): 21 - 25
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (354 KB) |  | HTML iconHTML  

    Land subsidence mainly caused by excessive extraction of groundwater is a growing and worldwide problem. Sustained decline in groundwater level has a direct impact on land surface elevation. However, the impact may not be consistent across subsidence areas. The spatially varying relationship between land subsidence and groundwater level variations remains unclear. This study explores the spatio-temporal changes based on the observed data of groundwater levels and benchmark elevations from 2002 to 2009 in the Choshuichi alluvial fan of central Taiwan and examines the spatial heterogeneity with geographically weighted regression (GWR). The results reveal that the occurrence and development of land subsidence is closely related to the groundwater pumping. Moreover, the influence of groundwater level on land subsidence is more significant in the inland area. The study can help to predict the land subsidence caused by the overdraft of groundwater and provide an explicit strategy for groundwater resource management. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A framework for spatial feature selection and scoping and its application to geo-targeting

    Publication Year: 2011 , Page(s): 26 - 31
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (409 KB) |  | HTML iconHTML  

    Predicting if a particular user clicks on a particular ad is of critical importance for internet advertising. Associations between Internet ad performance data, such as number of clicks or Click Through Rate, CTR, and demographic data may be very weak on the global level, but strong at the regional level. Identifying regions with strong associations of a continuous performance attribute with geo-features can create valuable knowledge for geo-targeted advertising. In this paper, we present a novel framework for interestingness scoping to identify such regions and discuss how such interestingness hotspots can be used for geo-feature evaluation with the goal to develop more accurate prediction models for advertisers. We also present the ZIPS algorithm that takes initial seed zip codes and discovers interestingness hotspots/coldspots, and a geo-feature preselection algorithm which automatically finds promising geo-features and identifies initial seed zipcodes for the ZIPS algorithm. We applied our framework to a large number of geo-spatial data sets, combining data from a major ad network, demographic data from the 2000 Census, and binary feature data from other sources. Our experimental results demonstrate that creating geo-features can double CTR performance for an Ad. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Comparisons with spatial autocorrelation and spatial association rule mining

    Publication Year: 2011 , Page(s): 32 - 37
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (485 KB) |  | HTML iconHTML  

    Spatial autocorrelation is a very general statistical property of economic variables, it indicates correlation of a variable with itself through space. Spatial association rule mining, discovery of interesting, meaningful rules in spatial databases, ignores autocorrelation of the spatial data, or just generalizes the spatial data into attribute data currently. In order to compare the results between spatial autocorrelation and spatial association rule mining, in this paper, the spatial association rules were mined by Apriori algorithm and it's development algorithm. Then, spatial autocorrelation analysis and spatial regression analysis were implemented on the same spatial data set. The experimental data is about the county-level revenue and population, education state, health state and social security state in China from 2000 to 2005. The results of the spatial association rules mining proves that economic level such as per capita revenue and social security have stronger correlation. The result of spatial autocorrelation is that from 2000 to 2005, national county-level per capita revenue, education, health and social security present positive spatial correlation. There is little interannual change in the spatial distribution of per capita revenue, and low economic level applies to almost all counties all over the nation. Education, the situation that low value areas are surrounded by high value areas universally exists, which shows that little significant positive influence from high level areas is exerted on low level areas. At the same time, the interprovincial education gap is gradually increasing. Health, in the year 2000 to 2005, there is a growing aggregation trend in China's county-level health spatial pattern, and there are more low value areas in health. Social security, in the research years, the aggregation trend is gradually decreasing. While spatial heterogeneity is increasing. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • 4D-GIS (4 dimensional GIS) as spatial-temporal data mining platform and its application to managementand monitoring of large-scale infrastructures

    Publication Year: 2011 , Page(s): 38 - 43
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (502 KB) |  | HTML iconHTML  

    A 4-dimensional geographic information system (4D-GIS) for use as a spatial-temporal data mining platform is introduced. The 4D-GIS effectively integrates, manages, and analyzes spatial-temporal data, defined as 4D data (2D, 3D, and time change data). In this paper, techniques for integrating and analyzing spatial-temporal data are provided in detail. First, the structures of 4D data and the essence of data integration are explained. A change difference management method for historical data and the homotopical representation of complicated 3D configurations are proposed. Second, the generative aspects of 4D data integration and the real-time monitoring of an integrated database are explained. The effectiveness of the system was verified by applying it to the large-scale infrastructure management of an elevated railway bridge and a sewer optical communication line. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An algorithm for clustering spatial lines based on connectivity for GML data

    Publication Year: 2011 , Page(s): 44 - 47
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (144 KB) |  | HTML iconHTML  

    Algorithm SLC is proposed for clustering spatial lines based on connectivity for GML data. At beginning the intersection relations of spatial lines between each other are computed, and then the connectivity of spatial lines is computed by intersection relations, finally spatial lines are clustered by algorithm K-means with a novel similarity measurement method based on the connectivity. The experimental results show the algorithm SLC is effective and efficient. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An extended ID3 decision tree algorithm for spatial data

    Publication Year: 2011 , Page(s): 48 - 53
    Cited by:  Papers (3)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (196 KB) |  | HTML iconHTML  

    Utilizing data mining tasks such as classification on spatial data is more complex than those on non-spatial data. It is because spatial data mining algorithms have to consider not only objects of interest itself but also neighbours of the objects in order to extract useful and interesting patterns. One of classification algorithms namely the ID3 algorithm which originally designed for a non-spatial dataset has been improved by other researchers in the previous work to construct a spatial decision tree from a spatial dataset containing polygon features only. The objective of this paper is to propose a new spatial decision tree algorithm based on the ID3 algorithm for discrete features represented in points, lines and polygons. As in the ID3 algorithm that use information gain in the attribute selection, the proposed algorithm uses the spatial information gain to choose the best splitting layer from a set of explanatory layers. The new formula for spatial information gain is proposed using spatial measures for point, line and polygon features. Empirical result demonstrates that the proposed algorithm can be used to join two spatial objects in constructing spatial decision trees on small spatial dataset. The proposed algorithm has been applied to the real spatial dataset consisting of point and polygon features. The result is a spatial decision tree with 138 leaves and the accuracy is 74.72%. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • QuCOM: K nearest features neighborhood based qualitative spatial co-location patterns mining algorithm

    Publication Year: 2011 , Page(s): 54 - 59
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (146 KB) |  | HTML iconHTML  

    Spatial Co-location patterns are similar to association rules but explore more relying spatial auto-correlation. They represent subsets of Boolean spatial features whose instances are often located in close geographic proximity. Existing co-location patterns mining researches only concern the spatial attributes, and few of them can handle the huge amount of non-spatial attributes in spatial datasets. Also, they use distance threshold to define spatial neighborhood. However, it is hard to decide the distance threshold for each spatial dataset without specific prior knowledge. Moreover, spatial datasets are not usually even distributed, so a unique distance value cannot fit an irregularly distributed spatial dataset well. Here, we proposed a qualitative spatial co-location pattern, which contains both spatial and non-spatial information. And the k nearest features (k-NF) neighbourhood relation was defined to set the spatial relation between different kinds of spatial features. The k-NF set of one feature's instances was used to evaluate close relationship to the other features. To find qualitative co-location patterns in large spatial datasets, some formal definitions were given, and a QuCOM (Qualitative spatial CO-location patterns Mining) algorithm was proposed. Experimental results on the USA thesis map data prove that QuCOM algorithm is accurate and efficient, and the patterns founded contain more interesting information. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • New approach for distributed clustering

    Publication Year: 2011 , Page(s): 60 - 65
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (286 KB) |  | HTML iconHTML  

    Nowadays the data collections are huge and in most cases do not reside in a centralised location. The latter complicates the task of traditional data mining techniques, as datasets are distributed and often heterogeneous. In this paper we propose a distributed approach based on the aggregation of models produced locally. The datasets will be processed locally on each node to produce clusters from local data then, we construct global clusters hierarchically. The aim of this approach is to minimise the communications, maximise the parallelism and load balance the work among different nodes of the system, and reduce the overhead due to extra processing while executing the hierarchical clustering. This technique is evaluated and compared to the sequential version using benchmark datasets and the results are very promising. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A new data mining method for early warning landslides based on parallel coordinate

    Publication Year: 2011 , Page(s): 66 - 70
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (179 KB) |  | HTML iconHTML  

    In this paper, a new data mining method is proposed on the basis of parallel coordinate for early warning of landslides. Landslides have resulted in many severe casualties and damaged structures and facilities. The proposed method is to analyse the landslide problems emerged with the parallel coordinates and its visualization function. It may simplify the establishment of complex model, and promote the visualization and analysis ability of spatial data, make closer relationship between spatial data and attribute data, and finally improve the effectiveness of landslide early warning. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Evaluation of a clustering technique based on game theory

    Publication Year: 2011 , Page(s): 71 - 76
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (195 KB) |  | HTML iconHTML  

    In this paper we focus on the task of clustering in data mining applications. We introduce a formulation of a new clustering algorithm by modelling the system as a cooperative game in strategic form using game theory. The goal is to partition a dataset into k clusters. Our approach has been applied to both simulated and real-world datasets. In addition, we have implemented functions based on the calculation of errors to track both similarity of the data within the same cluster and dissimilarity measure of the data elements between different clusters. Experimental results show that our algorithm is capable of providing a comprehensive description of the final solutions and it has good predictive capabilities. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Spatial regression analysis in hemorrhagic fever with renal syndrome (HFRS) in China

    Publication Year: 2011 , Page(s): 77 - 80
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (165 KB) |  | HTML iconHTML  

    With spatial regression analysis, the cause of the hemorrhagic fever with renal syndrome (HFRS) is detected in China in 2008. The results showed that: the incidence of HFRS is influenced by six factors: latitude, longitude of regional centers, the proportion of settlement place, the proportion of grassland, average elevation, and the average precipitation in October and November. Among these factors, the incidence is correlated significantly and negatively with the proportion of settlement place, while is associated significantly and positively with latitude and longitude of regional centres. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Evaluation of the spatio-temporal of soil salinity variation using data mining approach

    Publication Year: 2011 , Page(s): 81 - 86
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (588 KB) |  | HTML iconHTML  

    The issue of temporal and spatial variation in soil salinity is considered as a fundamental element in salinity monitoring. The aim of this study is to develop a framework which integrates image mining techniques with Fuzzy logic methodology to improve the evaluation of spatio-temporal variation of soil salinity in areas with lack of available ground observation. Intensity and duration of salinity was characterized in space by the deviation of the current NDVI at each location from its corresponding temporal mean value. Landsat and ASTER images data was used to provide frequent Normalized Difference Vegetation Index (NDVI) in cultivation phase for a period of 22 years. Evolution of salinity condition before planting season was assessed by applying stepwise regression method on image data for two available dataset. The regression equation was obtained between reflectance value of band three and the measured soil Electrical Conductivity (EC) from field. Validation of the developed algorithm was done by comparing the obtained outputs with 50 ground observations, available salinity reports, and previous soil salinity maps. The result revealed that the proposed framework can be considered as a cost and time effective tool for proper assessment of the spatio-temporal variation of soil salinity. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Water inrush dangerousness evaluation of coal floor from SDM

    Publication Year: 2011 , Page(s): 87 - 89
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (113 KB) |  | HTML iconHTML  

    The floor water inrush is a geology hazard in China. Tt is a complex nonequilibrium and nonlinear evolution process which is affected by many factors. Spatial Data mining (SDM) can reveal spatial patterns and features of water inrush. With the SDM technique, the water inrush dangerousness of 9# coal floor of zhangcun coal mine was evaluated. The result shows that the SDM technique is a feasible effective method to evaluate water inrush dangerousness of coal floor. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • TGCR: An efficient algorithm for mining swarm in trajectory databases

    Publication Year: 2011 , Page(s): 90 - 95
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (256 KB) |  | HTML iconHTML  

    Advance of positioning technology have enabled mass trajectory data of moving objects obtain more convenient. These moving objects always exists special behaviour correlation on spatio-temporal characteristics, and this information is important in some domains, such as prisoner monitoring, factory management, and the study of social behaviour. Many studies have focused on relative motion pattern mining algorithm, but the inefficiency of mining algorithms is still a problem. In this paper, we propose an efficient algorithm, Time Growth Cluster Recombinant algorithm (TGCR), for discovering swarm pattern, which is a group of relaxed aggregation moving objects. The algorithm construct maximum moving objectsets according to the clustering result of each timestamp, and record corresponding maximum time set of the maximum moving objectsets over time. TGCR employs three update rules to update candidate swarm list at each timestamp and proposes an insert rule to greatly reduce the redundant candidate items in the list. In addition, closure checking rule is presented for obtaining closed swarm patterns on fly. We performed an experimental evaluation of the correctness and efficiency of our algorithm using large synthetic data. The results of experiments demonstrate that TGCR discovers swarm patterns as same as objectGrowth algorithm and our algorithm have higher performance than objectGrowth. The further algorithm enhanced can be applicable to real-time trajectory data processing system. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A spatial data mining method for mineral resources potential assessment

    Publication Year: 2011 , Page(s): 96 - 99
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (277 KB) |  | HTML iconHTML  

    On the basis of multi-source geology spatial database and traditional spatial data mining, a spatial data mining method for mineral resources potential assessment was proposed in this paper, which the spatial characteristics and uncertainty of geology data were reasonable to consider. The method mainly include continuous geological spatial data discretization, spatial relationship abstracting and attribute transforming, mining metallogenic association rules and quality assessment, comprehensive evaluation of metallogenic association rules and potential assessment. Finally, the experiment of mineral potential assessment for iron deposits was performed in Eastern Kunlun, Qinghai province, China, using spatial data mining method and weights-of-evidence model, respectively. The results indicate that the prediction accuracy of spatial data mining was obvious higher than weights-of-evidence model's, the method is suitable for mineral resources potential assessment and its effectiveness is good. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Mining top-k closed co-location patterns

    Publication Year: 2011 , Page(s): 100 - 105
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (217 KB) |  | HTML iconHTML  

    In this paper, we present a problem to discover compact co-location patterns without minimum prevalence threshold. A spatial co-location is a set of spatial events being frequently observed together in nearby geographic space. A common framework for mining spatial co-location patterns employs a level-wised search method (like Apriori) to discover co-location patterns, and generates numerous redundant patterns since all of the 2l subsets of each length l event set the algorithms discover are included in the result set. In addition, most works of spatial co-location mining require the specification of a minimum prevalent threshold to find interesting co-location patterns. However, it is difficult for users to decide an appropriate threshold value without prior knowledge of their task-specific spatial data. To solve these problems, we propose a problem to mine top-k closed co-location patterns, where k is the desired number of patterns, and develop an algorithm to efficiently find the interesting patterns. The experiment result shows that the proposed algorithm is effective in computation. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • MDL-based segmentation of multi-attribute sequences

    Publication Year: 2011 , Page(s): 106 - 111
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (290 KB) |  | HTML iconHTML  

    Many real-life multi-attribute sequences (multi-sequences) have a segmental structure, with segments of differing structures of attribute dependencies, that reflect an evolving nature of the dependencies over time and space. We propose a new approach for discovering a segmental structure of such evolving dependencies in probabilistic terms as a sequence of Dynamic Bayesian Networks (DBN). We use the Minimum Description Length (MDL) Principle to partition the multi-sequence into non-overlapping and homogeneous segments by fitting an optimal sequence of DBNs to the multi-sequence. In experiments, conducted on daily rainfall data we showed the applicability of the method for discovering interesting spatio-temporal evolving dependencies between rainfall occurrences in south-western Australia. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The study on the exploratory spatial data mining method based on partial random walk and its application in GPS TEC analysis

    Publication Year: 2011 , Page(s): 112 - 115
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (194 KB) |  | HTML iconHTML  

    The Ionosphere plays an important role in atmosphere, whose globally distributed total electronic content (TEC) obtained by GPS technology is the important data source of geographic or earth information system for monitoring global change. This paper applies the rigging method of deionization variable theory to mine the knowledge of large scale of tendency variation and small scale of random variation, and discovered that the large scale tendency can be modelled as a 9 orders of globe harmony function, and the small scale variation more prefers to a zero mean non-stationary random process of symmetrically distributed. Applying the developed unit-root test, the small scale residual is identified with the characteristic of 3 orders of partial random walk, and thus the residuals after performing 3 orders of difference show the property of white noise process. The general Kriging predication method based on the partial random walk model is constructed to re-build the spatial process precisely. The result exhibits that the partial-random-walk-based test can be used to mine the auto-correlated structure of zero mean non-stationary error function or small scale variation, and the constructed general kriging method can improve the prediction result. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Discovering partial spatio-temporal co-occurrence patterns

    Publication Year: 2011 , Page(s): 116 - 120
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (129 KB) |  | HTML iconHTML  

    Spatio-temporal co-occurrence patterns represent subsets of object-types that are often located together in space and time. The aim of the discovery of partial spatio-temporal cooccurrence patterns (PACOPs) is to find co-occurrences of the object-types that are partially present in the database. Discovering PACOPs is an important problem with many applications such as discovering interactions between animals and identifying tactics in battlefields and games. However, mining PACOPs is computationally very expensive because the interest measures are computationally complex, databases are larger due to the archival history, and the set of candidate patterns is exponential in the number of object-types. Previous studies on discovering spatio-temporal co-occurrence patterns do not take into account the presence period (lifetime) of the objects in the database. In this paper, we define the problem of mining PACOPs, propose a new monotonic composite interest measure, and propose a novel PACOP mining algorithm. The experimental results show that the proposed algorithm is computationally more efficient than naïve alternatives. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • GeoKSGrid: A geographical knowledge grid with functions of spatial data mining and spatial decision

    Publication Year: 2011 , Page(s): 121 - 126
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (494 KB) |  | HTML iconHTML  

    Motivated by the lack of a geographical problem solving environment that is adequate to provide end users with reliable, open, distributed, and long lasting spatial data analyzing and knowledge discovery services, a novel geographical knowledge service platform - GeoKSGrid with functions of spatial decision support and distributed & parallel data mining is described in this paper. The basic concepts and state-of-the-art knowledge in grid research is discussed first. Then, the design of system architecture and the implementation of several most important modules of GeoKSGrid are illustrated. Finally, some demonstrative applications of the geographical knowledge services in real industry contexts is examined, combining with a brief interpretation of the processing results which confirm the practical value of the services and knowledge grid platform. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Extracting geographic knowledge from sensor intervention data using spatial association rules

    Publication Year: 2011 , Page(s): 127 - 130
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (180 KB) |  | HTML iconHTML  

    Large networks of sensors are used to detect intrusions and provide security at the borders of the United States. Sensor signals are used to detect possible intrusions such as illegal immigration traffic in drugs, weapons, and smuggled goods at specific targeted geographic locations. GIS systems can be used to capture, store and analyze this location based intervention data. Using a GIS system, a spatial database can be generated from the sensor intervention data which can take into account relevant geographic information in the vicinity of the sensed interventions. Important geographic features that are close to the intervention locations such as: plateaus, hills, valleys or roadways can be extracted and added to the analysis using ArcGIS. GIS techniques alone cannot reveal meaningful hidden information within geographic data. We have developed an integrated approach involving data mining and GIS techniques to extract patterns and trends in geographic data that can aid and inform analysis. Our approach uses both spatial and association data mining techniques. Spatial data mining is the process of discovering previously unknown, interesting and potentially useful patterns from spatial datasets. Applying association rule mining to the spatial data can reveal additional important spatial relationships and help determine the relevance and importance of the sensor data. Spatial association rule mining was used to discover patterns in the intervention data, such as linking a sensed intrusion with a potentially hidden location such as a canyon, to infer a high probability of illegal traffic or immigration. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.