The World-Wide Web presents new challenges to database researchers, especially in the area of query processing. Currently,
querying the World-Wide Web is done by using
Online Indices. These sites employ search engines, known as “robots”, that can scan the network periodically and form text based indices.
A severe limitation of these search services is that the structural information, namely the organization of documents into
parts pointing to each other, is lost. Several tasks, ranging from data mining to Intranet management, require the analysis
of the hypertext structural organization.
In this paper, we propose s simple graph based query language. In this language, both the query and its target are graphs.
We present and evaluate the efficiency of a general class of algorithms for answering graph queries. The algorithms’ definition
take into account two important facts of the WWW: (1) efficient algorithms must minimize the communication needed to answer
a query and (2) query evaluation involves a process of data graph exploration.