RDF is designed to provide a common way of describing resources so that it can be read and understood by computer applications. In RDF model, the statement in the resource description may correspond to a natural language statement, the resource corresponds to the subject in the natural language, the attribute type corresponds to the predicate, and the attribute value corresponds to the object. Meanwhile, RDF information has temporal attribute and spatial attribute. But classical RDF model can't show the spatial and temporal properties of resources. So, combining spatiotemporal information with RDF is necessary. However, SPARQL, the W3C-recommended query language of RDF, only meets the classic RDF queries. This chapter presents a spatiotemporal RDF representation model. Based on this model, a find isomorphic graphs of the query graph algorithm is introduced to obtain some candidate isomorphic graph of the query graph. Finally, the authors define the process of pattern matching.
Top1 Introduction
RDF(Broekstra et al, 2002) refers to the Resource Description Framework, which is recommended as the W3C standard in February 2004 (Klyne & Carroll, 2006). RDF is designed to provide a common way of describing information. In RDF model, a statement is a triple (Yan et al., 2008), which consists of resources, attribute types, and attribute values, indicating an attribute of the resource.
At present, spatiotemporal data has been widely used in all aspects of human life, such as traffic monitoring (Zhang et al., 2003), GIS (Mennis & Fountain, 2001), climate change monitoring (Antonić et al., 2001) and so on. Nowadays, data forms and dimensions are increasingly diverse. Data is no longer confined to one-dimensional and two-dimensional space, and more and more data not only show temporal attributes, but also show spatial attributes.
Therefore, introducing spatiotemporal attributes to RDF is the general trend(Tan & Yan, 2017; Kuper et al., 2016; Vatsavai et al., 2012; Venkateswara, 2012). Some related works were discussed. Koubarakis et al. (Koubarakis & Kyzirakos, 2010) develop the data model stRDF and the query language stSPARQL. Perry et al. (Perry et al., 2007) describe a framework built around the RDF metadata model for analysis of thematic, spatial and temporal relationships between named entities and present a set of semantic query operators. Perry et al. (Perry et al., 2011) present a SPARQL-ST query language, which is an extension of SPARQL for complex spatiotemporal queries, and describe an implementation of SPARQL-ST.
RDF graphs are composed of RDF triples. The growing popularity of RDF graph databases has generated some interesting algorithms for RDF graph queries(Tappolet & Bernstein, 2009; Nikitopoulos et al., 2019), such as: sub-graph matching, and pattern match. For example: Zou et al. (Zou et al., 2011) propose a graph-based approach to store and query RDF data, transform an RDF graph into a data signature graph, and develop a filtering rule for sub-graph query over the data signature graph. Li et al. (Li et al., 2019) develop an algorithm by extending graph homomorphism for efficiently evaluating sub-graph patterns over such fuzzy RDF graph. Zou et al. (Zou et al., 2009) transform the vertices into points in a vector space, and convert a pattern match query into a distance-based multi-way join problem over the converted vector space.
Among these, a pattern match query is more flexible compared to a sub-graph matching (Zou et al., 2009). But we have not yet found pattern matching and sub-graph matching related to spatiotemporal RDF.
In this chapter, according to the advantages of (Zhu et al., 2022) and (Li et al., 2019) mentioned above, we propose pattern matching query based on spatiotemporal RDF. The main research contents are divided into the following aspects:
- •
a representation model of spatiotemporal RDF.
- •
an algorithm to obtain some candidate isomorphic graph of the query graph.
- •
a pattern matching process of spatiotemporal RDF graph.
The reminder of the chapter is organized as follows: Section 2 introduces a modeling process of spatiotemporal RDF. Section 3 describes three steps for querying spatiotemporal RDF graph. Section 4 discusses the correctness and validity of our algorithm. And Section 5 gives conclusions and directions for future work.