SL WANG
580 California St., Suite 400
San Francisco, CA, 94104
Software and applications of spatial data mining
https://doi.org/10.1002/WIDM.1180Uploaded (2016)
visibility
…
description
3 pages
link
1 file
Sign up for access to the world's latest research
checkGet notified about relevant papers
checkSave papers to use in your research
checkJoin the discussion with peers
checkTrack your impact
Abstract
Most big data are spatially referenced, and spatial data mining (SDM) is the key to the value of big data. In this paper, SDM are overviewed in the aspects of software and application. First, spatial data are summarized on their rapid growth, distinct characteristics, and implicit values. Second, the principles of SDM are briefed with the descriptive definition, fundamental attributes, discovery mechanism , and usable methods. Third, SDM software is presented in the context of software components, developing methodology, typical software for geographical information system (GIS) data and remote sensing (RS) images, and software trend. Fourth, SDM applications are outlined on GIS data, RS image, and spatio-temporal video data. The final is the concluding remarks and perspectives.
Related papers
A Survey of Spatial Data Mining Methods Databases and Statistics Point of Views
Data Warehousing and Web Engineering
This chapter reviews the data mining methods that are combined with Geographic Information Systems (GIS) for carrying out spatial analysis of geographic data. We will first look at data mining functions as applied to such data and then highlight their specificity compared with their application to classical data. We will go on to describe the research that is currently going on in this area, pointing out that there are two approaches: the first comes from learning on spatial databases, while the second is based on spatial statistics. We will conclude by discussing the main differences between these two approaches and the elements they have in common.
A Brief on Spatial Data Mining
International Journal on Computer Science and Engineering
Spatial data mining is the process of mining the spatial and spatiotemporal data, in order to discover otherwise hidden and unknown patterns and trends in data. Its application areas include business prospecting, store prospecting, hospital prospecting and automobile insurance etc. This paper includes a survey of spatial data mining, its types, techniques and roles in the field of research. Clustering, Classification, Cloropeth display have been the main focus of the paper. All kinds of clustering such as partitioning based, hierarchical based, grid based, model based and density based clustering techniques have been applied here. The outcome of the survey is a consolidated study of all the above techniques and methods of spatial data mining including a detailed summary, advantages, disadvantages, comparison and distinguishes among several techniques. Such type of survey has not been existing till date. So, our main objective behind the paper was to bring a standardized and uniform platform for all the techniques of spatial data mining.
Spatial data mining and geographic knowledge discovery—An introduction
Computers, Environment and Urban Systems, 2009
Voluminous geographic data have been, and continue to be, collected with modern data acquisition techniques such as global positioning systems (GPS), high-resolution remote sensing, location-aware services and surveys, and internet-based volunteered geographic information. There is an urgent need for effective and efficient methods to extract unknown and unexpected information from spatial data sets of unprecedentedly large size, high dimensionality, and complexity. To address these challenges, spatial data mining and geographic knowledge discovery has emerged as an active research field, focusing on the development of theory, methodology, and practice for the extraction of useful information and knowledge from massive and complex spatial databases. This paper highlights recent theoretical and applied research in spatial data mining and knowledge discovery. We first briefly review the literature on several common spatial data-mining tasks, including spatial classification and prediction; spatial association rule mining; spatial cluster analysis; and geovisualization. The articles included in this special issue contribute to spatial data mining research by developing new techniques for point pattern analysis, prediction in space-time data, and analysis of moving object data, as well as by demonstrating applications of genetic algorithms for optimization in the context of image classification and spatial interpolation. The papers concludes with some thoughts on the contribution of spatial data mining and geographic knowledge discovery to geographic information sciences.
Spatial Data Mining on Remote Sensing Perspective
2016
Spatial data mining has been evolved as a special branch of data mining since the last two decades as the conventional data mining algorithms fail to produce good results from large spatial databases. Satellite images carry rich information of the geographical coverage area, which can be harnessed using efficient algorithms that consider spatial aspects into, while processing. Several algorithms, that are build up on the state-of-the–art techniques such as AI, ANN, SVM, Fuzzy logic, and are categorized into the classical grouping of clustering, classification and frequent pattern mining have grown dimensionally and have stretched into multidisciplinary research areas. In this letter, we will discuss the various efforts taken in the last two decades in the field of spatial data mining on the perspective of satellite/ remote sensing images.
A survey on data mining and knowledge discovery techniques for spatial data
International Journal of Business Information Systems, 2015
Spatial data in databases, specifically in geographic information systems (GIS) data centres, can produce huge amounts of data. This great volume of information includes useful and non-useful data along with some hidden patterns which can be better managed and controlled by utilising some novel artificial intelligence methods. In this research we have a brief survey on the application of statistical methods like data mining and knowledge discovery on spatial data. First, we present an overview of geospatial data mining and knowledge discovery techniques, including spatial clustering, classification, prediction, associate rules and pattern analysis. Then, some challenges faced by geographic knowledge discovery in geographic information systems (GIS) have discussed in order to have a more clear idea of the future's researches in this area.
Comparative Study of Spatial Data Mining Techniques
International Journal of Computer Applications, 2015
Spatial data mining is a mining knowledge from large amounts of spatial data. Spatial data mining algorithms can be separated into four general categories: clustering and outlier detection, association and co-location method, trend detection and classification. All these methods have been compared according to various attributes. This paper introduces the fundamental concepts of widely known spatial data mining algorithms in a comparative way. It focuses on techniques and their unique features.
Prominence of Spatial Database in Geographical Information Systems
2014
Spatial Database is a collection of spatially referenced data that acts as a model of reality. Geographical Information System includes spatial data which is either in raster or vector form and data derived from remote sensors are increasingly utilized as source for GIS. Spatial data is the core part of the Geographical Information System. Data mining techniques are studied to discover knowledge from GIS database and remote sensing image data in order to improve land use or land cover classification. Remote sensing (RS) technologies was utilized to extract some of the important spatially variable parameters, such as land cover and land use (LCLU), from satellite images. Land Use Land Cover (LULC) classification from remotely sensed data is an important research and widely used in remote sensing application. The goal of this paper is to provide broad overview of Geographical information system and remote sensing data which includes spatial data. Keywords—Spatial data, Geographical In...
Spatiotemporal Data Mining in the Era of Big Spatial Data: Algorithms and Applications
Proceedings ACM SIGSPATIAL Workshop on Analytics for Big Geospatial Data (BigSpatial – 2012), 2012
Spatial data mining is the process of discovering interesting and previously unknown, but potentially useful patterns from the spatial and spatiotemporal data. However, explosive growth in the spatial and spatiotemporal data, and the emergence of social media and location sensing technologies emphasize the need for developing new and computationally efficient methods tailored for analyzing big data. In this paper, we review major spatial data mining algorithms by closely looking at the computational and I/O requirements and allude to few applications dealing with big spatial data.
Spatial Data Infrastructures and Data Mining - An Introduction
2008
Spatial data infrastructures (SDIs) and data mining have been a key challenge of a modern society. SDIs as well as spatial data mining should become a topic of the annual conference EnviroInfo. The main objectives of our paper are: (1) To provide a short overview about spatial data infrastructures (definitions, goals, hierarchy of SDIs, and relations to other initiatives). (2) To describe the three main data mining techniques which have been often used for spatial datasets. (3) To report on the development of a spatial database at the Leibniz Institute of Ecological and Regional Development (IOER) for the key research area “Settlement and Open Space Development Monitor” and to illustrate some applications of the spatial data mining in selected IOER projects such as “Research to Ecological and Economic Performance Potentials of 116 German Regional Cities.
Overview
Software and applications of
spatial data mining
Deren Li,1 Shuliang Wang,1,2* Hanning Yuan2* and Deyi Li3
Most big data are spatially referenced, and spatial data mining (SDM) is the key
to the value of big data. In this paper, SDM are overviewed in the aspects of soft-
ware and application. First, spatial data are summarized on their rapid growth,
distinct characteristics, and implicit values. Second, the principles of SDM are
briefed with the descriptive definition, fundamental attributes, discovery mech-
anism, and usable methods. Third, SDM software is presented in the context of
software components, developing methodology, typical software for geographical
information system (GIS) data and remote sensing (RS) images, and software
trend. Fourth, SDM applications are outlined on GIS data, RS image, and spatio-
temporal video data. The final is the concluding remarks and perspectives. © 2016
John Wiley & Sons, Ltd
How to cite this article:
WIREs Data Mining Knowl Discov 2016, 6:84–114. doi: 10.1002/widm.1180
INTRODUCTION and veracity of spatial data.2,3 Internet-based geospa-
tial community of volunteered geographical informa-
S patial data play an import role in the sustainable
development of natural resources and human
society. By using data, human civilization has gone
tion4 and location-based sensed service, e.g., call
logs, mobile-banking transactions, online user-
generated content such as blog posts and Tweets,
through preliminary sensing the world, memorizing online searches, and satellite images, make spatial
the instances, recording the writings, inheriting the relationships more and more complex. The research
history, exchanging the information, communicating from International Data Corporation5 has shown
with each other, and uncovering the rules. Spatial that as of 2003, humans have created a total of 5EB
data are closely related to human daily life and per- data, while in the year 2011, the amount of data that
meated all walks of businesses. Most data are spa- had been copied and produced has exceeded 1.8ZB.
tially referenced,1 which help people understand the It is expected that by 2020, global data usage will
real world through the information world on the reach 35.2ZB, which needs 37.6 billion hard drives
basis of locations. The rapid developments of instru- of 1TB capacity to store. ‘WorldView’ satellite2 has a
ments, techniques ,and infrastructures to capture spa- spatial resolution of 0.5 m, and its direct georeferen-
tial data, e.g., global positioning systems (GPS), cing accuracy can reach 2–3 m. The Earth
remote sensing (RS), geographical information sys- Observing-1 (EO-1) satellite has 220 imaging spec-
tem (GIS), smart planet, and the network of things, trums, and the spectral range is 400–2500 nm, and it
have greatly prompted the volume, variety, velocity, can gain the spectral resolution of 10 nm.
These spatial data are voluminous and grow
rapidly, which have far exceeded the human ability
*Correspondence to: [email protected], [email protected]
1
to timely interpret and use with common
State Key Laboratory of Information Engineering in Surveying,
Mapping and Remote Sensing, Wuhan University, Wuhan, China
techniques.1,2,4,6–30 For example, a newspaper pub-
2 lished the same article of the same author on two dif-
School of Software, Beijing Institute of Technology, Beijing,
China ferent pages of ‘legal community’ and ‘youth topics.’
3
Department of Computer Science and Technology, Tsinghua Uni- Another newspaper published three articles in the
versity, Beijing, China edition of ‘home appliances,’ ‘lifestyle,’ and ‘science
Conflict of interest: The authors have declared no conflicts of inter-
and technology,’ all to compare video compact disc
est for this article. (VCD), China video disc (CVD), and digital versatile
84 © 2016 John Wiley & Sons, Ltd Volume 6, May/June 2016
WIREs Data Mining and Knowledge Discovery Software and applications of SDM
F I G U R E 1 4 | Chinese HFT map of 2008 with 1 km × 1 km resolution.
FI GU RE 15 | The nighttime light monthly composites: (a) March 2011; and (b) February 2014.
1. Analysis of abnormal motion trajectory. analysis, feature blocks, or feature point analy-
Extract the linear trajectory and characteristics; sis (Figure 16).
identify and classify the characteristics. Solve 3. Classification combines with pattern recogni-
the problem of trajectory cross or separation tion. For example, using SVM, Ann, Boost
and so on. Solve the problem of analyzing the rapid classification technology to classify and
multiobjective overlapping anomaly. identify moving target.
2. Target analysis based on the hybrid model, 4. Spatiotemporal analysis for the video from
including color model analysis, shape model multicamera (Figure 17). For the target at the
Volume 6, May/June 2016 © 2016 John Wiley & Sons, Ltd 105
Overview wires.wiley.com/dmkd
same time, due to the fixed location of camera, popular techniques and products of spatiotemporal
the times that targets arise are interconnected; video data mining.
with this association information, we can com-
bine image target-matching technology to real-
ize the temporal and spatial correlation Disaster Forecast and Management
analysis. In recent years, people have successfully utilized
5. The camera video analysis in different periods. advanced information technologies to forecast disas-
Extraction is not dependent on the time-related ter events timely with high accuracy by assessing
information of image characteristics such as environmental conditions and predicting events (such
environment light, contrast ratio, and other as forest fires, population growth, and other natural
information. Retrieve and recognize pedestrians or human development issues). In particular, the use
or vehicles appear in different periods of satellite navigation and positioning technology can
automatically. pinpoint the location of the disaster. The discovered
knowledge applied in a specific industry, specific
In this way, we can also delete large amounts scenes, and specific solution can better support deci-
of videos recording people’s normal activities and sion making and action1,3. For example, after the
private activities that need to be protected and only Hurricane Katrina disaster in Louisiana and Missis-
keep the data about suspicious cars and people as sippi, when local electric power, telecommunications,
well as data about people whom need to be cared for roads, and other local infrastructure were destroyed,
(dementia in the elderly and mentally retarded chil- the US Coast Guard helicopter used GPS to locate
dren). Because of the technical difficulty, there are no and rescue the victims. In the 2010 earthquake in
Yushu, the unmanned aerial vehicle (UAV) images
with a resolution of 0.2 m were used for distribution
planning, traffic jams monitoring, and disaster asses-
sing. Comparing the high-resolution images of Galle
city railway station captured before and after the tsu-
nami may help interpret the damage of buildings.
Spatial mitigation satellites, RS satellites, com-
munication, and navigation satellites have been
widely used in disaster management, such as earth-
quakes, tsunamis, typhoons (hurricanes), floods,
droughts, geological disasters, and fire. Currently,
the Disaster Monitoring Constellation is a system
that consists of satellites from various countries
through international cooperation. This system has a
high time resolution, large monitor range, and fast
response. For disaster mitigation and relief, IBM
Sahana Rescuers-Centered Disaster Reduction sys-
FI G UR E 1 6 | Pedestrian classification based on hog operator. tem, with the collaboration of personnel,
FI GU RE 17 | Space-time video sequence-based event detection.
106 © 2016 John Wiley & Sons, Ltd Volume 6, May/June 2016