Regular Description of Context-free Graph Languages

Abstract

A set of (labeled) graphs can be defined by a regular tree language and one regular string language for each possible edge label, as follows. For each tree t from the regular tree language the graph gr(t) has the same nodes as t (with the same labels), and there is an edge with label : from node x to node y if the string of labels of the nodes on the shortest path from x to y in t belongs to the regular string language for :. Slightly generalizing this definition scheme, we allow gr(t) to have only those nodes of t that have certain labels, and we allow a relabeling of these nodes. It is shown that in this way exactly the class of C-edNCE graph languages (generated by C-edNCE graph grammars) is obtained, one of the largest known classes of context-free graph languages. ]

Formalization of Link Farm Structure Using Graph Grammar

“…Since the structure of a graph affects the complexity of graph grammars, different kinds of graph grammar rules are proposed such as context-free graph grammars [11] and context-sensitive graph grammars [8]. In this work, we consider context-free graph grammars.…”

“…Graph grammars provide a mechanism by which the generation and transformation can be performed by a graph language [11]. Generally, graph grammars are difficult to construct and require considerable effort by experts.…”