Karin Harbusch

Combining deictic gestures and natural language for referent identification

In virtually all current natural-language dialog systems, users can only refer to objexts by using linguistic descriptions. However, in human face-to-face conversation, participants frequently use various sorts of deictic gestures as well. In this paper, we will present the referent identification component of XTRA, a system for a natural-language access to expert systems. XTRA allows the user to combine NL input together with pointing gestures on the terminal screen in order to refer to objects on the display. Information about the location and type of this deictic gesture, as well as about the linguistic description of the referred object, the case frame, and the dialog memory are utilized for identifying the object. The system is tolerant in respect to impreciseness of both the deictic and the natural language input. The user can thereby refer to objects more easily, avoid referential failures, and employ vague everyday terms instead of precise technical notions.

XTRA: a natural-language access system to expert systems

Abstract The XTRA access system to expert systems is presented which is aimed at rendering the interaction with expert systems easier for inexperienced users. XTRA communicates with the user in a natural language (German), extracts data relevant to the expert system from his/her natural-language input, answers user queries as to terminology and provides user-accommodated natural-language verbalizations of results and explanations provided by the expert system. A number of novel artificial intelligence techniques have been employed in the development of the system, including the combination of natural-language user input and user gestures on the terminal screen, referent identification with the aid of four different knowledge sources, simultaneous communication of the access system with the user and the expert system, fusion of two complementary knowledge bases into a single one, and the design of a natural-language generation component which allows for a controlled interaction between the “what-to-say” and the “how-to-say” parts to yield a more natural output. XTRA is being developed independently of any specific expert system. In its first application the access to an expert system in the income tax domain is being realized.

Performance Grammar: A declarative definition

In this paper we present a definition of Performance Grammar (PG), a psycholinguistically motivated syntax formalism, in declarative terms. PG aims not only at describing and explaining intuitive judgments and other data concerning the well–formedness of sentences of a language, but also at contributing to accounts of syntactic processing phenomena observable in language comprehension and language production. We highlight two general properties of human sentence generation, incrementality and late linearization, which make special demands on the design of grammar formalisms claiming psychological plausibility. In order to meet these demands, PG generates syntactic structures in a two-stage process. In the first and most important ‘hierarchical’ stage, unordered hierarchical structures (‘mobiles’) are assembled out of lexical building blocks. The key operation at work here is typed feature unification, which also delimits the positional options of the syntactic constituents in terms of so-called topological features. The second, much simpler stage takes care of arranging the branches of the mobile from left to right by ‘reading–out’ one positional option of every constituent.In this paper we concentrate on the structure assembly formalism in PG’s hierarchical component. We provide a declarative definition couched in an HPSG–style notation based on typed feature unification. Our emphasis throughout is on linear order constraints.

Towards an adaptive communication aid with text input from ambiguous keyboards

Ambiguous keyboards provide efficient typing with low motor demands. In our project1 concerning the development of a communication aid, we emphasize adaptation with respect to the sensory input. At the same time, we wish to impose individualized language models on the text determination process. UKO--II is an open architecture based on the Emacs text editor with a server/client interface for adaptive language models. Not only the group of motor impaired people but also users of watch--sized devices can profit from this ambiguous typing.

A quantitative model of word order and movement in English, Dutch and German complement constructions

We present a quantitative model of word order and movement constraints that enables a simple and uniform treatment of a seemingly heterogeneous collection of linear order phenomena in English, Dutch and German complement constructions (Wh-extraction, clause union, extraposition, verb clustering, particle movement, etc.). Underlying the scheme are central assumptions of the psycholinguistically motivated Performance Grammar (PG). Here we describe this formalism in declarative terms based on typed feature unification. PG allows a homogenous treatment of both the within- and between-language variations of the ordering phenomena under discussion, which reduce to different settings of a small number of quantitative parameters.

Incremental Syntax Generation with Tree Adjoining Grammars

With the increasing capacity of AI systems the design of human-computer interfaces has become a favorite research topic in AI. In this paper we focus on aspects of the output of a computer. The architecture of a sentence generation component — embedded in the WIP system — is described. The main emphasis is laid on the motivation for the incremental style of processing and the encoding of adequate linguistic units as rules of a Lexicalized Tree Adjoining Grammar with Unification.

AN EFFICIENT PARSING ALGORITHM FOR TREE ADJOINING GRAMMARS

In the literature, Tree Adjoining Grammars (TAGs) are propagated to be adequate for natural language description --- analysis as well as generation. In this paper we concentrate on the direction of analysis. Especially important for an implementation of that task is how efficiently this can be done, i.e., how readily the word problem can be solved for TAGs. Up to now, a parser with O(n6) steps in the worst case was known where n is the length of the input string. In this paper, the result is improved to O(n4 log n) as a new lowest upper bound. The paper demonstrates how local interpretion of TAG trees allows this reduction.

Generating Clausal Coordinate Ellipsis Multilingually: A Uniform Approach Based on Postediting

Present-day sentence generators are often incapable of producing a wide variety of well-formed elliptical versions of coordinated clauses, in particular, of combined elliptical phenomena (Gapping, Forward and Backward Conjunction Reduction, etc.). The applicability of the various types of clausal coordinate ellipsis (CCE) presupposes detailed comparisons of the syntactic properties of the coordinated clauses. These nonlocal comparisons argue against approaches based on local rules that treat CCE structures as special cases of clausal coordination. We advocate an alternative approach where CCE rules take the form of postediting rules applicable to nonelliptical structures. The advantage is not only a higher level of modularity but also applicability to languages belonging to different language families. We describe a language-neutral module (called Elleipo; implemented in JAVA) that generates as output all major CCE versions of coordinated clauses. Elleipo takes as input linearly ordered nonelliptical coordinated clauses annotated with lexical identity and coreferentiality relationships between words and word groups in the conjuncts. We demonstrate the feasibility of a single set of postediting rules that attains multilingual coverage.

The Sentence Fairy: a natural-language generation system to support children's essay writing

We built an NLP system implementing a ‘virtual writing conference’ for elementary-school children, with German as the target language. Currently, state-of-the-art computer support for writing tasks is restricted to multiple-choice questions or quizzes because automatic parsing of the often ambiguous and fragmentary texts produced by pupils presents insurmountable problems. Here, we follow a different course by deploying natural language generation technology to evaluate and improve the grammatical quality of student output. Based on an abstract representation of the story under construction, all paraphrases of simple and combined clauses are generated fully automatically. From this source, the system produces exercises enabling the pupils to improve their sentences. We apply parsing technology only in the teacher mode, where new stories are entered into the system in a simple manner. The system we describe here is a prototype including a full exercise generation mode and a rudimentary teacher mode.

ELLEIPO: a module that computes coordinative ellipsis for language generators that don't

Many current sentence generators lack the ability to compute elliptical versions of coordinated clauses in accordance with the rules for Gapping, Forward and Backward Conjunction Reduction, and SGF (Subject Gap in clauses with Finite/Fronted verb). We describe a module (implemented in JAVA, with German and Dutch as target languages) that takes non-elliptical coordinated clauses as input and returns all reduced versions licensed by coordinative ellipsis. It is loosely based on a new psycholinguistic theory of coordinative ellipsis proposed by Kempen. In this theory, coordinative ellipsis is not supposed to result from the application of declarative grammar rules for clause formation but from a procedural component that interacts with the sentence generator and may block the overt expression of certain constituents.