Many probabilistic models for natural language are now written in terms of hierarchical tree structure. Tree-based modeling still lacks many of the standard tools taken for granted in (finite-state) string-based modeling. The theory of tree transducer automata provides a possible framework to draw on, as it has been worked out in an extensive literature. We motivate the use of tree transducers for natural language and address the training problem for probabilistic tree-totree and tree-to-string transducers.

Extended top-down tree transducers (transducteurs generalises descendants [Arnold, Dauchet: Bi-transductions de forets. ICALP'76. Edinburgh University Press. 1976]) received renewed interest in the field of Natural Language Processing. Here those transducers are extensively and systematically studied. Their main properties are identified and their relation to classical top-down tree transducers is exactly characterized. The obtained properties completely explain the Hasse diagram of the induced classes of tree transformations. In addition, it is shown that most interesting classes of transformations computed by extended top-down tree transducers are not closed under composition.

We consider the verification of non-recursive C programs manipulating dynamic linked data structures with possibly several next pointer selectors and with finite domain non-pointer data. We aim at checking basic memory consistency properties (no null pointer assignments, etc.) and shape invariants whose violation can be expressed in an existential fragment of a first order logic over graphs. We formalise this fragment as a logic for specifying bad memory patterns whose formulae may be translated to testers written in C that can be attached to the program, thus reducing the verification problem considered to checking reachability of an error control line. We encode configurations of programs, which are essentially shape graphs, in an original way as extended tree automata and we represent program statements by tree transducers. Then, we use the abstract regular tree model checking framework for a fully automated verification. The method has been implemented and successfully applied on several case studies.

. The finiteness of ranges of tree transductions is shown to be decidable for TBY + , the composition closure of macro tree transductions. Furthermore, TBY + definable sets and TBY + computable relations are considered, which are obtained by viewing a tree as an expression that denotes an element of a given algebra. A sufficient condition on the considered algebra is formulated under which the finiteness problem is decidable for TBY + definable sets and for the ranges of TBY + computable relations. The obtained result applies in particular to the class of string languages that can be defined by TBY + transductions via the yield mapping. This is a large class which is proved to form a substitution-closed full AFL. 1 Introduction The finiteness problem is one of the classical decidability problems in formal language theory. For a given language of interest, one usually does not wish to know whether that language is finite (because it usually is not), but rather whether the l...

Unfortunately, the class of transformations computed by linear extended top-down tree transducers with regular look-ahead is not closed under composition. It is shown that the class of transformations computed by certain linear bimorphisms coincides with the previously mentioned class. Moreover, it is demonstrated that every linear epsilon-free extended top-down tree transducer with regular look-ahead can be implemented by a linear multi bottom-up tree transducer. The class of transformations computed by the latter device is shown to be closed under composition, and to be included in the composition of the class of transformations computed by top-down tree transducers with itself. More precisely, it constitutes the composition closure of the class of transformations computed by nite-copying top-down tree transducers.

. A software system called Treebag---Tree-Based Generator---is presented. The aim behind Treebag is to allow for the generation and visualization of objects of all kinds: pictures, trees, graphs, strings, numbers, etc. The basic principle is that tree generators like, for instance, regular tree grammars, generate terms over symbols that are interpreted by appropriate algebras as operations on the domain of objects under consideration. Thus, every term is viewed as an expression that denotes one of the objects of interest. These objects can be visualized using appropriate displays. 1 Introduction In mathematics and computer science, one of the most natural and universal ways to denote elements of a given data space is to use expressions over a suitable set of operations, like p 3=7 + 5. The usefulness of this concept is even strengthened by the fact that such an expression may contain variables---in which case it does not denote a particular, single object, but a set of objects or a ...

. A characterization of the sets of hypertrees generated by hyperedgereplacement graph grammars is given. The characterization says that these sets are exactly those which have the form val(T ), where T , a set of terms over hyperedgereplacement operations, is the output language of a finite-copying top-down tree transducer. Furthermore, the terms in T may be required to consist of hyperedgereplacement operations whose underlying hypergraphs are hypertrees. The result is closely related to a similar characterization that was obtained for the case of string graphs by Engelfriet and Heyker some years ago. In fact, the results of this paper also yield a new proof for the characterization by Engelfriet and Heyker. 1 Introduction Hyperedge-replacement graph grammars, also called context-free hypergraph grammars, are well-studied devices for the generation of graph and hypergraph languages (see, e.g., [Hab92, Eng97, DHK97]). Their basic operation is the replacement of a non-terminal hypered...

Tree series transformations computed by bottom-up and top-down tree series transducers are called bottom-up and top-down tree series transformations, respectively. (Functional) compositions of such transformations are investigated. It turns out that the class of bottomup tree series transformations over a commutative and complete semiring is closed under left-composition with linear bottom-up tree series transformations and right-composition with boolean deterministic bottom-up tree series transformations. Moreover, it is shown that the class of top-down tree series transformations over a commutative and complete semiring is closed under right-composition with linear, nondeleting top-down tree series transformations. Finally, the composition of a boolean, deterministic, total top-down tree series transformation with a linear top-down tree series transformation is shown to be a top-down tree series transformation.

. The notion of P-interpreted top-down tree generators is introduced, combining the nondeterministic nature of grammars as known from formal language theory with the innite renement of pictures studied in fractal geometry. 1 Introduction This paper introduces an approach to generate languages of fractals, thus combining fractal geometry with one of the main ideas of formal language theory, namely to consider innite sets of objects related by a common grammatical description. The theory of formal languages is normally concerned with the grammatical generation of sets of discrete objects, these sets being called languages. Traditional questions ask for the properties of these languages and their recognition by various sorts of automata. While the objects of interest had initially been words, the ideas were soon taken up by researchers interested in more complex data structures like arrays, trees, and graphs (see the recently published books [RS97] for a representative overview of...

During the execution of functional logic programs, E-unification problems have to be solved quite frequently, where the underlying equational theory is induced by recursive functions. But, what about the decidability of those E-unification problems? Up to now, there does not exist a concrete answer to this question for classes of equational theories which are induced by particular recursive functions. In this paper, we try to give an answer to this question by drawing and verifying a strict border between undecidability and decidability of E-unification problems for particular classes of recursive functions. Since this result shows that the E-unification problem is undecidable even for a very restricted class of recursive functions, the nondeterministic implementations of those problems in functional logic programming languages are justified.