Abstract. This paper studies the RDF model from a database perspective. From this point of view it is compared with other database models, particularly with graph database models, which are very close in motivations and use cases to RDF. We concentrate on query languages, analyze current RDF trends, and propose the incorporation to RDF query languages of primitives which are not present today, based on the experience and techniques of graph database research. 1

this paper we report upon a graph-based approach to such an integration. Our use of graphs has two key advantages : firstly, graphs are formally defined, well-understood structures; secondly, it is widely accepted that graph-based formalisms considerably enhance the usability of complex systems [19]. Graphs have been used in conjunction with a number of conventional data models, for example the hierarchical and network models [35], the entity-relationship model [9] and a recent extension thereof for complex objects [27], and various semantic data models [16, 20, 31]. Graphs or hypergraphs [6] have also been used more recently in [12, 17, 23, 25, 33, 36] as a data modelling tool in their own right. We give a comparison between this recent work and our own approach in Section 4 of the paper. Directed graphs have also been the foundation of Hypertext databases [11, 33]. Such databases are graphs consisting of nodes which refer to units of stored information (typically text) and of named links. Each link connects two nodes, the "source" and the "destination". Links are traversed either forwards (from source to destination) or backwards (from destination to source). The process of traversing named links and examining the text associated with nodes is called

Graph database models can be characterized as those where data structures for the schema and instances are modeled as graphs or generalizations of them, and data manipulation is expressed by graph-oriented operations and type constructors. These models flourished in the eighties and early nineties in parallel to object oriented models and their influence gradually faded with the emergence of other database models, particularly the geographical, spatial, semistructured and XML. Recently, the need to manage information with inherent graph-like nature has brought back the relevance of the area. In fact, a whole new wave of applications for graph databases emerged with the development of huge networks (e.g. Web, geographical systems, transportation, telephones), and families of networks generated due to the automation of the process of data gathering (e.g. social and biological networks). The main objective of this survey is to present in a single place the work that has been done in the area of graph database modeling, concentrating in data structures, query languages and integrity constraints.

Currently database researchers are investigating new data models in order to remedy the deficiences of the flat relational model when applied to non-business applications. Herein we concentrate on a recent graph-based data model called the hypernode model. The single underlying data structure of this model is the hypernode which is a digraph with a unique defining label. We present in detail the three components of the model, namely its data structure, the hypernode, its query and update language, called HNQL, and its provision for enforcing integrity constraints. We first demonstrate that the said data model is a natural candidate for formalising hypertext. We then compare it with other graph-based data models and with set-based data models. We also investigate the expressive power of HNQL. Finally, using the hypernode model as a paradigm for graph-based data modelling, we show how to bridge the gap between graph-based and set-based data models, and at what computational cost this can...

With more and more large networks becoming available, mining and querying such networks are increasingly important tasks which are not being supported by database models and querying languages. This paper wants to alleviate this situation by proposing a data model and a query language for facilitating the analysis of networks. Key features include support for executing external tools on the networks, flexible contexts on the network each resulting in a different graph, primitives for querying subgraphs (including paths) and transforming graphs. The data model provides for a closure property, in which the output of every query can be stored in the database and used for further querying.

We have developed a graph-based data model called the Hypernode Model whose single data structure is the hypernode, a directed graph whose nodes may themselves reference further directed graphs. A prototype database system supporting this model is being developed at London University as part of a project whose aims are threefold : (i) to ascertain the expressiveness and flexibility of the hypernode model, (ii) to experiment with various querying paradigms for this model, and (iii) to investigate the suitability of the directed graph as a data structure supported throughout all levels of the implementation. The purpose of this paper is to report upon our findings to date. 1.

An important step in data analysis is the exploration of data. For traditional relational databases one of the most powerful tools for performing such analysis is the relational database and the aggregates and rankings that they can compute: for instance, simple statistics such as the average number of links between two types of entities (relations) are easily computed using a query on a relational database and may already provide valuable information. However, for the exploration of graph data, relational databases may not be most practical and scalable. For instance, a statistic such as the shortest path between two given nodes cannot be computed by a relational database. Surprisingly, however, tools for querying graph and network databases are much less well developed than for relational data, and only recently an increasing number of studies are devoted to graph or network databases. Our position is that the development of such graph databases is important both to make basic graph mining easier and to prepare data for more complex types of analysis. An important component of such databases is the language that is used to enable aggregating queries, such as shortest path queries.In this paper, we propose an extension to a previously proposed query language. This extension allows for querying and analyzing databases by using aggregates and ranking. A notable feature of our language is that it also supports probabilistic graph queries by conceiving of such queries as aggregating queries. We demonstrate its value on a simple data analysis task.

Abstract not found