Flexible Query of Uncertain Spatiotemporal XML Data

1 Introduction

With the continuous development and popularization of science and technology, the users of spatiotemporal databases have gradually changed, and a large number of ordinary users can access spatiotemporal databases and obtain information through various channels. Most ordinary users do not know the actual situation of the data, often can not accurately define the query conditions. Therefore, users want to access the database in a flexible query way.

In recent years, with the rapid growth of spatiotemporal data, the demand for spatiotemporal data modeling and query is increasing. For spatiotemporal data modeling, Chen et al. (Chen et al., 2015) proposed a spatiotemporal data model based on compressed linear reference, which transformed network temporal geographical entities in three-dimensional space into two-dimensional compressed linear reference space. Cheng et al. (Cheng et al., 2021) proposed a dynamic spatiotemporal logic, which is used to represent and reason dynamic spatiotemporal knowledge. Guan et al. (Guan et al., 2017) proposed a trajectory index method, which extended GeoHash algorithm to meet the requirements of high-frequency update and common trajectory query operations, thus accelerating the time-consuming spatiotemporal query of massive trajectory data. What’s more, spatiotemporal query technology mainly includes the following aspects: Simple Temporal Query (Hao, 2011) mainly studies how to query the state of an object at a certain time. Furthermore, Ma et al. (Ma, 2012) proposes selective query, which mainly queries all spatiotemporal data objects passing through a certain area within a certain time interval. On this basis, join query (Li, 2012) is proposed, which queries objects that are at a certain distance from a certain data object in a certain time interval. Que et al. (Que et al.,2014) proposes spatiotemporal nearest neighbor query, which queries the data object closest to an object or region within a certain time interval.

Because XML has good self-description and extensibility, people have made some efforts in spatiotemporal data modeling and query relaxation based on XML. Some researchers have proposed an adaptive query relaxation method based on XML (Liu et al., 2010). This method returns only the results related to the query results. Dfgdfg et al. (Dfgdfg et al., 2011) put forward XPath query relaxation method. This method uses rewriting rules to relax query conditions. Bai et al. (Bai et al., 2016) used XQuery to study spatiotemporal operations, and studied query result processing by listing query examples. In addition, Bai et al. (Bai et al., 2021) has made some achievements in modeling and querying fuzzy or uncertain spatiotemporal data based on XML.

Due to the uncertainty of real information, most of them use fuzzy sets and fuzzy logic form to model language for query. The main idea of this work is to extend the SQL language and add an additional classical DBMS layer to evaluate fuzzy predicates ((Kandel et al., 1984); (Galindo et al., 2006); (Mama et al., 2019); (Mama et al., 2019); (Mama et al., 2019)). In addition, Meng et al. (Meng et al., 2009) proposed a query relaxation method based on semantic similarity to solve the problem that the results returned by Web database queries are empty or the number of results returned is small. Based on another idea, the query relaxation method with context preference (Yan et al., 2012) is generated to meet the user's personalized query needs. This method stores context preference by constructing interest tree. Although the research on flexible query technology of Web database has developed to a certain extent, the related research on flexible query of uncertain spatiotemporal data is still lacking.

In this chapter, we propose a flexible query method of uncertain Spatiotemporal XML Data, and give the corresponding ranking method of results. Based on this model, this chapter proposes a T-List structure which can quickly locate nodes that meet the time range of query conditions, and reduces the time required for query matching. This chapter proposes a method of relaxing initial query and general attribute query on uncertain spatiotemporal data model and a flexible query method of relaxing spatiotemporal attribute on T-List structure. Single-Relaxation algorithm is proposed for single attribute relaxation, and Multiple-Relaxation algorithm is proposed for combined attribute relaxation. In order to facilitate users to view the query results, this chapter proposes a RSort algorithm to sort the exact results and the extended results.