Artificial Intelligence, 63 (1-2), 69-142

This paper presents PARSECa system for generating connectionist parsing networks from example parses. PARSEC is not based on formal grammar systems and is geared toward spoken language tasks. PARSEC networks exhibit three strengths important for application to speech processing: 1) they learn to parse, and generalize well compared to handcoded grammars; 2) they tolerate several types of noise; 3) they can learn to use multi-modal input. Presented are the PARSEC architecture and performance analyses along several dimensions that demonstrate PARSEC's features. PARSEC's performance is compared to that of traditional grammar-based parsing systems.

This paper presents PARSECa system for generating connectionist parsing networks from example parses. PARSEC is not based on formal grammar systems and is geared toward spoken language tasks. PARSEC networks exhibit three strengths important for application to speech processing: 1) they learn to parse, and generalize well compared to handcoded grammars; 2) they tolerate several types of noise; 3) they can learn to use multi-modal input. Presented are the PARSEC architecture and performance analyses along several dimensions that demonstrate PARSEC's features. PARSEC's performance is compared to that of traditional grammar-based parsing systems.

This paper presents PARSECa system for generating connectionist parsing networks from example parses. PARSEC is not based on formal grammar systems and is geared toward spoken language tasks. PARSEC networks exhibit three strengths important for application to speech processing: 1)they learn to parse, and generalize well compared to handcoded grammars; 2) they tolerate several types of noise; 3) they can learn to use multi-modal input. Presented are the PARSEC architecture and performance analyses along several dimensions that demonstrate PARSEC's features. PARSEC's performance is compared to that of traditional grammar-basedparsing systems. 1 INTRODUCTION While a great deal of research has been done developing parsers for natural language, adequate solutionsfor some of the particular problems involved in spoken language have not been found. Among the unsolved problems are the difficulty in constructing task-specific grammars, lack of tolerance to noisy input, and inability to effectively utilize non-symbolic information.This paper describes PARSECa system for generating connectionist parsing networks from example parses.

Phrase-structure grammars are an effective representation for important syntactic and semantic aspects of natural languages, but are computationally too demanding for use as language models in real-time speech recognition. An algorithm is described that computes finite-state approximations for context-free grammars and equivalent augmented phrase-structure grammar formalisms. The approximation is exact for certain context-free grammars generating regular languages, including all left-linear and right-linear context-free grammars. The algorithm has been used to construct finite-state language models for limited-domain speech recognition tasks.

We present a novel, modular, recurrent connectionist network architecture which learns to robustly perform incremental parsing of complex sentences. From sequential input, one word at a time, our networks learn to do semantic role assignment, noun phrase attachment, and clause structure recognition for sentences with passive constructions and center embedded clauses. The networks make syntactic and semantic predictions at every point in time, and previous predictions are revised as expectations are affirmed or violated with the arrival of new information. Our networks induce their own "grammar rules" for dynamically transforming an input sequence of words into a syntactic/semantic interpretation. These networks generalize and display tolerance to input which has been corrupted in ways common in spoken language.

We present a novel, modular, recurrent connectionist network architecture which learns to robustly perform incremental parsing of complex sentences. From sequential input, one word at a time, our networks learn to do semantic role assignment, noun phrase attachment, and clause structure recognition for sentences with passive constructions and center embedded clauses. The networks make syntactic and semantic predictions at every point in time, and previous predictions are revised as expectations are affirmed or violated with the arrival of new information. Our networks induce their own "grammar rules" for dynamically transforming an input sequence of words into a syntactic/semantic interpretation. These networks generalize and display tolerance to input which has been corrupted in ways common in spoken language.

We present a novel, modular, recurrent connectionist network architecture of complexwhich learns to robustly perform incremental parsing sentences. From sequential input, one word at a time, our networks learn to do semantic role assignment, noun phrase attachment, and clause structure recognition for sentences with passive constructions and center embedded clauses. The networks make syntactic and semantic predictions at every point in time, and previous predictions are revised as expectations are affirmed or violated with the arrival of new information. Our networks induce their own "grammar rules" for dynamically transforming an input sequence of words into a syntactic/semantic interpretation.

The Workshop on Future Directions in NLP was held at Bolt Beranek and Newman, Inc. (BBN), in Cambridge, Massachusetts, from 29 November to 1 December 1989. The workshop was organized and hosted by Madeleine Bates and Ralph Weischedel of the BBN Speech and Natural Language Department and sponsored by BBN's Science Development Program.

Many barriers exist today that prevent effective industrial exploitation of current and future AI research. These barriers can only be removed by people who are working at the scientific forefront in AI and know potential industrial needs. The Knowledge Processing Laboratory's research and development concentrates in the following areas: (1) natural language interfaces to knowledge-based systems and databases; (2) theoretical and experimental work on qualitative modeling and nonmonotonic reasoning for future knowledge-based systems; (3) application-specific language design, in particular, Prolog extensions; and (4) desi gn and analysis of neural networks. This article gives the reader an overview of the main topics currently being pursued in each of these areas.