This paper describes an efficient and robust implementation of a bidirectional, head-driven parser for constraint-based grammars. This parser is developed for the OVIS system: a Dutch spoken dialogue system in which information about public transport can be obtained by telephone. After a Review

Alfa-informatica & BCN, University of Groningen {vannoord, gosse}Olet. rug. nl parsers and generators for natural languages. The package is written in Sicstus Prolog and Tcl/Tk. The system provides a graphical user interface with a command interpreter, and a number of visualisation tools, including visualisation of fea- ture structures, syntax trees, type hierarchies, lexical hierarchies, feature structure trees, definite clause definitions, grammar rules, lexical entries, and graphs of statis- tical information of various kinds.

The NWO Priority Programme Language and Speech Technology is a 5-year research programme aiming at the development of spoken language information systems. In the Programme, two alternative natural language processing (NLP) modules are developed in parallel: a grammar-based (conventional, rule-based) module and a data-oriented (memorybased, stochastic, DOP) module. In order to compare the NLP modules, a formal evaluation has been carried out three years after the start of the Programme. This paper describes the evaluation procedure and the evaluation results. The grammar-based component performs much better than the data-oriented one in this comparison.

this report we refer to this update as the `best update').

Abstract not found

In this paper we will describe ongoing work on an application of Ginzburg's account for short answers in order to improve the nlp component of a spoken dialogue system by using contextual information for disambiguating speech input. 1 Introduction In this paper we will address the question of how to incorporate contextual information for improving the Natural Language Processing (nlp) component of a spoken dialogue system. In the ovis project we are working on a spoken dialogue system for public transport information. Some of the characteristics of most of these types of systems are: 1) The system operates on a very restricted domain, 2) the dialogues are mostly simple and short, 3) The user is mainly restricted to answering system questions, 4) The user might answer with full sentences, but will in practice mostly provide short answers. For the ovis system we developed a grammar that on the one hand covers most important grammatical constructions in Dutch and that on the other han...

he user utterance. In such cases, a linguistic analysis component might proceed by computing a partial analysis of the input. NLU needs to be robust for three reasons. Firstly, it is quite di#cult to anticipate in the grammar all linguistic constructions that might occur. This is one of the traditional problems for grammar-based NLU. Secondly, spoken language is full of hesitations, corrections, false starts etc. which are not always easy to detect. The third reason is that the utterance that was actually spoken is not guaranteed to be a path in the word graph, due to limitations of state-ofthe -art speech recognition, as well as external factors such as background noise, etc. These observations indicate that robustness and disambiguation are two very important problems to be solved in the NLU component. Given the nature of the proposed application, it will be clear that the system is supposed to run in real-time, i.e. it is not supposed to leave the user waiting for the requested in