As an increasing volume of data objects in the real world has spatial features and temporal features, spatiotemporal data emerge in more and more spatiotemporal applications. Although keyword query has been studied on spatiotemporal data in traditional database, relatively little work has been carried out to investigate keyword query of uncertain spatiotemporal data. In this chapter, the authors investigate uncertain spatiotemporal data representation based on XML. Building upon this, they propose a keyword query approach for uncertain spatiotemporal XML data, which includes the data structure of a dynamic keyword data warehouse, probability calculation, keyword query algorithm, as well as pruning and sorting algorithms for keyword query. Finally, the experimental results demonstrate the performance advantages of this approach. Such approach optimizes the structure of spatiotemporal XML data, which facilitates its keyword query.
Top1 Introduction
Recently, with the explosive growth of spatiotemporal data, the demand for spatiotemporal data modeling and querying is increasing. For spatiotemporal data modeling, Chen et al. (Chen et al., 2016) propose a spatiotemporal data model based on a compressed linear reference to transform network time geographic entities in 3D space to 2D compressed linear reference space. The advantage of this model is that spatiotemporal operations and index structures can be utilized to implement spatiotemporal operations and queries directly. The event-driven spatiotemporal data model (Li et al., 2014) is proposed to dynamically express and simulate the spatiotemporal processes of spatiotemporal phenomena. Cheng et al. (Cheng et al., 2021) propose a dynamic spatiotemporal logic for representing and reasoning dynamic spatiotemporal knowledge. For spatiotemporal data querying, Kim et al, (Kim et al., 2000) present a spatiotemporal data model that supports a bitemporal concept for spatial objects and designs a spatiotemporal database query language denoted as STQL. Guan et al. (Guan et al., 2017) present a trajectory indexing method to accelerate time-consuming spatiotemporal queries of massive trajectory data by extending GeoHash algorithm to satisfy the requirements for both high-frequent updates and common trajectory query operations.
At the same time, XML has good self-descriptiveness and extensibility and has become an international standard for representation and exchange in the Web (An et al., 2005). Therefore, some efforts have been made in modeling and querying spatiotemporal data based on XML (Yuan et al., 2010). Zipf and Krüger (Zipf & Krüger, 2001) propose a self-consistent framework to describe temporal data by an XML schema, which is combined with GML schema to realize spatiotemporal XML schema. Liu and Wan (Liu & Wan, 2010) use XML to describe a feature-based spatiotemporal data model, which can use Native XML Database to store the spatiotemporal data. Chen and Revesz (Chen & Revesz, 2003) study a query language for spatiotemporal data based on XML, which supports spatiotemporal data heterogeneity at the language level. Bai et al. (Bai et al., 2016) study spatiotemporal operations using XQuery, and investigate query result processing by listing query examples. In addition to those spatiotemporal data modeling and querying approaches based on XML, there are also several further achievements on fuzzy or uncertain spatiotemporal data modeling and querying based on XML (Jeung et al., 2008; Cheng & Wang, 2008; Boulila et al., 2011; Bai et al., 2015; Xu et al., 2016; Ma et al., 2018; Bai et al., 2018; Chen et al., 2018; Bai et al., 2018; Bai et al., 2017; Bai et al., 2021).