Spatiotemporal data mining: a survey on challenges and open problems

Spatiotemporal data mining is the core research topic of geographic information science. In the era of big data, the explosive growth of geographic spatiotemporal data puts forward an urgent demand for spatiotemporal knowledge discovery, which promotes the continuous development of spatiotemporal data mining technology. However, the universal heterogeneity and sparse distribution characteristics of spatiotemporal big data restrict the realization of spatiotemporal data mining algorithms, and significantly affect the description and analysis capabilities of natural and social complex systems. Thus, this paper focuses on the series of bottlenecks faced in the expression and application of heterogeneous and sparsely distributed spatiotemporal data. We systematically summarized the research status and existing problems of several key spatiotemporal mining tasks including missing spatiotemporal data interpolation, sparse spatiotemporal data reconstruction, and spatiotemporal state prediction, condensed four key scientific problems, and gave four corresponding solutions. The proposed methods are expected to enrich the method system in the field of spatiotemporal data mining and improve the quality and application value of spatiotemporal data modeling.

Spatio-temporal association data refers to spatial feature information flows that have spatio-temporal associations with each other and change over time. The mining of spatio-temporal related data has attracted much attention and has been widely used in many fields. At present, many researches of spatio-temporal data mining can only infer the association mode of spatio-temporal objects from the system. Therefore, based on the complex network theory, this paper proposed a spatio-temporal data networked node correlation inference method based on Bayesian method. Based on the network topology structure, the sample statistics method was designed according to time and space information. The conditional probability between network nodes was calculated so that it was applied to the node correlation causal inference. Finally, the method was applied to two typical spatio-temporal correlation data networks - seismic network and pest-relational network, and the prediction accuracy was 76% and 80%, respectively. In addition, this paper analyzed the relationship between the accuracy of the prediction and other factors such as test set size, structural core and magnitude limits.

Detecting anomalies in spatio-temporal flow data by constructing dynamic neighbourhoods

With the recent advances in IoT and other smart devices, an explosive amount of data, both spatially and temporally significant are generated. Discovering interesting or useful patterns from these spatiotemporal data is referred to as spatiotemporal data mining. These patterns could be unordered, totally ordered or partially ordered based on the temporal ordering. This work focusses on the totally ordered patterns or sequential patterns from spatiotemporal event database. Spatiotemporal event sequence miner finds sequence of events that overlaps spatially and temporally. Traditional approaches discover patterns that are frequent in the entire dataset. In this work a clustering-assisted approach to find regionally or locally frequent spatiotemporal pattern is proposed. The proposed Clustering assisted Regional Spatiotemporal Event Sequence (CReST) mining approach overcomes the bias caused by uneven distribution of spatiotemporal events while mining patterns. Chicago crime dataset is used for evaluating the proposed approach with traditional sequence mining algorithm.

Spatio-temporal data mining is an emerging area with increasing importance in a variety of applications, such as homeland security, mobile services, surveillance systems, and health monitoring applications. However, mining in spatio-temporal databases is still in its infancy. Existing work on spatio-temporal data mining has mainly focused on three types of patterns: evolution patterns of natural phenomena, frequent movements of objects over time, and space-time clusters. While there has been much research on association rule mining on transactional, spatial, and temporal data, there is little literature on finding interesting associations in spatio-temporal data. In this chapter, we introduce the early attempts at spatio-temporal data mining and review the techniques to discover various interesting spatio-temporal patterns. This is followed by a review of the traditional association rules mining algorithms and their variants on transactional data, temporal data, and spatial data.

This study investigated how to manage and use the spatio-temporal data during land use. To realize spatio-temporal data mining, the key issue to be solved in technology is how to organize spatio-temporal data. The spatio-temporal data model is just generated to solve this problem. According to analysis of spatio-temporal data changing character, two factors are very important in spatio-temporal data model. One is object, and the other is event. The object is an essential element that comprises spatio-temporal data information. The event is a fundamental factor that result in changing spatio-temporal data. A new spatio-temporal data model based on both object and event (OESTDM) was put forward to meet the demand. How to organize data, build relations between objects and storage data in spatio-temporal database are also emphatically discussed. At last, a dynamic land use information system based on OESTDM is presented. The system realized the function of managing spatio-temporal data, the retrospect of history information, tracing land parcel change, Area-Data-Change-Statistic and Analysis, and so on efficiently, by taking the land parcel data of some regions in Zhejiang province for study.

As the number, volume and resolution of spatio-temporal datasets increases, traditional statistical methods for dealing with such data are becoming overwhelmed. Nevertheless, the spatio-temporal data are rich sources of information and knowledge, waiting to be discovered. The field of spatio-temporal data mining (STDM) emerged out of a need to create effective and efficient techniques in order to turn the massive data into meaningful information and knowledge. This chapter reviews the state of the art in STDM research and applications, with emphasis placed on three key areas, including spatio-temporal prediction and forecasting, spatio-temporal clustering and spatio-temporal visualization. The future direction and research challenges of STDM are discussed at the end of this chapter.

The pervasiveness of GPS-equipped mobile devices has been nurturing an unprecedented amount of semanticsrich spatiotemporal data. The confluence of spatiotemporal and semantic information offers new opportunities for extracting valuable knowledge about people's behaviors, but meanwhile also introduces its unique challenges that render conventional spatiotemporal data mining techniques inadequate. Consequently, mining semantics-rich spatiotemporal data has attracted significant research attention from the data mining community in the past few years. In this tutorial, we start with reviewing classic spatiotemporal data mining tasks and identifying the new opportunities introduced by semantics-rich spatiotemporal data. Subsequently, we provide a comprehensive introduction of existing techniques for mining semantics-rich spatiotemporal data, covering topics including spatiotemporal activity mining, spatiotemporal event discovery, and spatiotemporal mobility modeling. Finally, we discuss about the limitations of existing research and identify several important future directions.

The recent advances in technologies for collecting spatial and spatio-temporal data (e.g. Cellular Phones, GPS devices) has facilitated the collection of data referenced in space and time. These huge collections of data often hide interesting information which conventional systems are unable to discover. Spatial and spatio-temporal data require complex data preprocessing, transformation, data mining, and post-processing techniques to extract novel, useful, and understandable patterns. The importance of spatial data mining is growing with the increasing incidence and importance of large geo-spatial datasets such as maps, repositories of remote-sensing images, trajectories of moving objects generated by mobile devices, etc. Applications include Mobile-commerce industry (location-based services), climatological effects of El Nino, land-use classification and global change using satellite imagery, finding crime hot spots, and local instability in traffic. The main goal of this tutorial is to disseminate this research field in Brazil, giving an overview of the current state of the art and the main methodologies and algorithms for spatial and spatiotemporal data mining.

A considerable amount of data emerging in spatiotemporal applications requires much attention on managing spatiotemporal data both from academia and industry. On the other hand, XML has several excellent characteristics due to its simplicity, readability, and portability. However, although XML has been employed to model and handle spatiotemporal data, the study of spatiotemporal XML data has only recently started and still merits further attention. In this paper, we study spatiotemporal data model for sea surface meteorological data in XML, and then propose marking dictionaries for sea surface meteorological data in XML. The feature of our approach is that our spatiotemporal data model is application-oriented. Moreover, such approach could provide consolidations of application-oriented spatiotemporal models.

Abstract. With the wide application of Big Data, Artificial Intelligence and Internet of Things in geographic information technology and industry, geospatial big data arises at the historic moment. In addition to the traditional "5V" characteristics of big data, which are Volume, Velocity, Variety, Veracity and Valuable, geospatial big data also has the characteristics of "Location Attribute". At present, the study of geospatial big data are mainly concentrated in: knowledge mining and discovery of geospatial data, Spatiotemporal big data mining, the impact of geospatial big data on visualization, social perception and smart city, geospatial big data services for government decision-making support four aspects. Based on the connotation and extension of geospatial big data, this paper comprehensively defines geospatial big data comprehensively. The application of geospatial big data in location visualization, industrial thematic geographic information comprehensive service and geographic data science and knowledge service is introduced in detail. Furthermore, the key technologies and design indicators of the National Geospatial Big Data Platform are elaborated from the perspectives of infrastructure, functional requirements and non-functional requirements, and the design and application of the National Geospatial Public Service Big Data Platform are illustrated. The challenges and opportunities of geospatial big data are discussed from the perspectives of open resource sharing, management decision support and data security. Finally, the development trend and direction of geospatial big data are summarized and prospected, so as to build a high-quality geospatial big data platform and play a greater role in social public application services and administrative management decision-making.

Locating the radio signal coverage hole (SCH) and signal weak coverage area (SWCA) in the mobile network and taking appropriate actions to improve network quality are the major tasks for mobile network operators (MNO). A novel approach, Signal Coverage Hole and weak coverageArea DIscovering Model (SCHADIM), is proposed based on spatiotemporal data mining on the raw data collecting from location based service (LBS) to achieve this goal. It reuses the communication raw data in LBS-based applications and transforms it into mobile network communication quality monitoring information by integrating the GIS (geographical information system), road network database and mobile network base station database. By this way, the vehicles in the LBS-based applications can be regarded as the mobile network signal probing vehicles, which provides plentiful information for discovering the SCH/SWCA. Comparing to the traditionalmobile network signal probing vehicle method, which is known as radio site verification (RSV) method, the proposed SCHADIM has the spatiotemporal coverage as well as cost advantages. A mobile network monitoring system, cellular network quality safeguard (CQS), has been implemented based on the proposed model which combines with the domain expert heuristics to provide decision support information for optimizing the mobile network.

Data Warehouse (DW) is topic-oriented, integrated, static datasets which are used to support decision-making. Driven by the constraint of mass spatio-temporal data management and application, Spatio-Temporal Data Warehouse (STDW) was put forward, and many researchers scattered all over the world focused their energy on it.Although the research on STDW is going in depth , there are still many key difficulties to be solved, such as the design principle, system framework, spatio-temporal data model (STDM), spatio-temporal data process (STDP), spatial data mining (SDM) and etc. In this paper, the concept of STDW is discussed, and analyzes the organization model of spatio-temporal data. Based on the above, a framework of STDW is composed of data layer, management layer and application layer. The functions of STDW should include data analysis besides data process and data storage. When users apply certain kind of data services, STDW identifies the right data by metadata management system, then start data processing tool to form a data product which serves the data mining and OLAP. All varieties of distributed databases (DDBs) make up data sources of STDW, including Digital Elevation Model (DEM), Diagnosis-Related Group (DRG), Data Locator Group (DLG), Data Objects Management (DOM), Place Name and other databases in existence. The management layer implements heterogeneous data processing, metadata management and spatio-temporal data storage. The application layer provides data products service, multidimensional data cube, data mining tools and on-line analytical process.

Visualization, Mining, and Density Analyses of Spatial and Spatio-Temporal Data

Availability of location-based services and mobile computing enabled the collection of spatial and temporal data. Spatiotemporal database integrates spatial and temporal aspects of data. The objective of this chapter is to offer conceptual clarity and state of the art technologies and tools for spatiotemporal data engineering. This chapter discusses spatiotemporal data representation, modeling, and database design, indexing, analysis, and mining for developing the applications. It also covers spatiotemporal big data processing systems.