Description

One of the goals of the Semantic Web is to create a web of data. Data are connected and may be pretty complex. They can be represented naturally as a graph. The expressiveness of graph models in representing data and knowledge is well-known. Therefore, graph databases have earned much attention for a couple of years. There is a claim that graph databases are better than RDF (Resource Description Framework) stores in representing graph-oriented data. Representing graphs-oriented data is a challenge not only for RDF stores but also for the RDF data model.

The advantage of RDF is that it is a part of standards Semantic Web technology. Structured data dominated by the relational model (RDB) cannot be ignored as the resource to populate the Semantic Web. Unfortunately, existing mapping approaches do not consider the semantics of RDB in the real world. A few converting approaches from the relational model to the graphs model must comprehensively consider this semantic issue. The other work, which maps graph models to RDF, also has some limitations. Firstly, this book explains the Semantic Mapping from the relational model to the graph model without losing the semantics, especially the mapping of links. The result of this approach is a directed property hypergraph, which is closer to the semantics of the real world.

Next, by considering a directed property hypergraph, this book proposes an attributed predicate RDF (AP-RDF), including the mapping approach from the relational model. AP-RDF augments the role of predicates in handling the complexity of data. It adopts the concept of property hyperedges, which can have attributes. Hence, predicates can have attributes as well. The examples show that all explained approaches are promising and give some contributions to the representation and the query technology layers of the Semantic Web.