Abstract In this paper, we continue the research on the power of contextual grammars with selection languages from subfamilies of the family of regular languages. In the past, two independent hierarchies have been obtained for external and internal contextual grammars, one based on selection languages defined by structural properties (finite, monoidal, nilpotent, combinational, definite, ordered, non-counting, power-separating, suffix-closed, commutative, circular, or union-free languages), the other one based on selection languages defined by resources (number of non-terminal symbols, production rules, or states needed for generating or accepting them). In a previous paper, the language families of these hierarchies for external contextual grammars were compared and the hierarchies merged. In the present paper, we compare the language families of these hierarchies for internal contextual grammars and merge these hierarchies.

In this paper, we continue the research on the power of contextual grammars with selection languages from subfamilies of the family of regular languages. In the past, two independent hierarchies have been obtained for external and internal contextual grammars, one based on selection languages defined by structural properties (finite, monoidal, nilpotent, combinational, definite, ordered, non-counting, power-separating, suffix-closed, commutative, circular, or union-free languages), the other one based on selection languages defined by resources (number of non-terminal symbols, production rules, or states needed for generating or accepting them). In the present paper, we compare the language families of these hierarchies for external contextual grammars and merge the hierarchies.

A contextual grammar is a language generating mechanism inspired by generating sentences in natural languages. An existing string can be extended to a new string of the language by adjoining a context before and behind the string or by inserting it into the string around some subword. The first mode is called external derivation whereas the second mode is called internal derivation. If conditions are given, around which words which contexts can be adjoined, we speak about contextual grammars with selection. We give an overview about the generative capacity of contextual grammars (working externally or internally) where the selection languages belong to subregular language classes. All languages generated by contextual grammars where all selection languages are elements of a certain subregular language family form again a language family. We compare such families which are based on finite, monoidal, nilpotent, combinational, definite, suffix-closed, ordered, commutative, circular, non-counting, power-separating, or union-free languages, or based on languages defined by restrictions regarding the descriptional complexity.

In this paper, we approach the problem of accepting all recursively enumerable languages by accepting networks of evolutionary processors (ANEPs, for short) with a fixed architecture. More precisely, we show that every recursively enumerable language can be accepted by an ANEP with an underlying graph in the form of a star with 13 nodes or by an ANEP with an underlying grid with 13 × 4 = 52 nodes as well as by ANEPs having underlying graphs in the form of a chain, a ring, or a wheel with 29 nodes each. In all these cases, the size and form as well as the general working strategy of the constructed networks do not depend on the accepted language; only the rewriting rules and the filters associated to each node of the networks depend on this language. Noteworthy is also the fact that the filtering process is implemented using random context conditions only. Our results answer problems which were left open in a paper published by J. Dassow and F. Manea at the conference on Descriptional Complexity of Formal Systems (DCFS) 2010 and improve a result published by B. Truthe at the conference on Non-Classical Models of Automata and Applications (NCMA) 2013.

In this paper, we study networks of evolutionary processors where the filters are chosen as special regular sets. We consider networks where all the filters belong to a set of languages that are accepted by deterministic finite automata with a fixed number of states. We show that if the number of states is bounded by two, then every recursively enumerable language can be generated by such a network. If the number of states is bounded by one, then not all regular languages but non-context-free languages can be generated.

We extend the edit operators of substitution, deletion, and insertion of a symbol over a word by introducing two new operators (partial copy and partial elimination) inspired by biological gene duplication. We define a disruption measure for an opera

In this paper, a new definition of accepting networks – called target based accepting networks – is given. In a target based accepting network of evolutionary processors, each node is equipped with a target condition. As soon as a node contains a wor

We discuss the power of networks of evolutionary processors where only two types of nodes are allowed. We prove that (up to an intersection with a monoid) every recursively enumerable language can be generated by a network with one deletion and one insertion node. Networks with an arbitrary number of deletion and substitution nodes only produce finite languages, and for each finite language one deletion node or one substitution node is sufficient. Networks with an arbitrary number of insertion and substitution nodes only generate context-sensitive languages, and (up to an intersection with a monoid) every context-sensitive language can be generated by a network with one substitution node and one insertion node. All results are optimal with respect to the number of nodes.