ABSTRACT
Graph-Structured Data (GSD) are abundant today in areas such as communication and traffic networks, web information systems, digital libraries, social networks, biological data management, and so on. GSD is easy to deal with as there is no need to worry about making the data fit rigid relational tables. They can be constructed on the fly using underlying data sources that we already have in hand. They can also be easily merged to create bigger databases, once they agree in a common node referencing scheme. Also, there are ongoing, enormous efforts for capturing the immense knowledge available on the Web and representing it in the form of “subject-predicate-object” triples, which are in essence “node-label-node” graph edges. The graphs they form are the well-known RDF graphs (often called triplestores) and they are immensely popular today (cf. [8,16]). Semistructured data (SSD) is another example of graph-structured data. Querying and reasoning on SSD has been the theme of a very extensive and multifaceted research for the last decade.
