Donald E. Walker

Natural Language Access To A Melanoma Data Base

This paper describes ongoing research towards developing a system that will allow physicians personal accesa to patient medical data through natural language queries to support both patient management and clinical research. A prototype system has been implemented for a small data base on malignant melanoma. The physician can input queries in English that retrieve specified data for particular patients or for groups of patients satisfying certain characteristics, that perform simple calculations, that allow browsing through the data base, and that assist in identifying relations among attributes. The system supports dialogue interactions; that is, the user can follow a line of inquiry to test a particular hypothesis by entering a sequence of queries that depend on each other. Classes of questions that can be processed are described and examples using the system are given.

Current Issues in Computational Linguistics: In Honour of Don Walker

Introduction. Donald Walker: a Remembrance. Section 1: The Task of Natural Language Processing. Natural Language Processing: an Historical Review K. Sparck Jones. On Getting a Computer to Listen J. Robinson. Utterance and Objective: Issues in Natural Language Communication B. Grosz. On the Proper Place of Semantics in Machine Translation M. King. Developing a Natural Language Interface to Complex Data G.G. Hendrix, E.D. Sacerdoti. User-Needs Analysis and Design Methodology for an Automated Document Generator K. Kukich, K. McKeown, J. Shaw, J. Robin, J. Lim, N. Morgan, J. Philips. Section 2: Building Computational Lexicons. Machine-Readable Dictionaries and Computational Linguistics Research B. Boguraev. Research Toward the Development of a Lexical Knowledge Base for Natural Language Processing R.A. Amsler. Discovering Relationships Among Word Senses R.J. Byrd. Machine Readable Dictionary as a Source of Grammatical Information E. Hajicova, A. Rosen. The ITT Lexical Database: Dream and Reality S. Pin-Ngern Conlon. Visions of the Digital Library: Views on Using Computational Linguistics and Semantic Nets in Information Retrieval J.L. Klavans. Anatomy of a Verb Entry: from Linguistic Theory to Lexicographic Practice B.T. Atkins, J. Kegl, B. Levin. Issues for Lexicon Building N. Calzolari. Outline of a Model for Lexical Databases N. Ide, J. le Maitre, J. Veronts. Construction-Based MT Lexicons L. Levin, S. Nirenburg. Dependency-Based Grammatical Information in the Lexicon P. Sgall. Semantics in the Brains Lexicon -- Some Preliminary Remarks on its Epistemology H. Schnelle. Section 3: The Acquisition and Use of LargeCorpora. The Ecology of Language D.E. Walker. Representativeness in Corpus Design D. Biber. The Text Encoding Initiative C.M. Sperberg-McQueen. Discrimination Decisions for 100,000 Dimensional W.A. Gale, K.W. Church, D. Yarowsky. Acquisition and Exploitation of Textual Resources for NLP S. Armstrong-Warwick. The Center for Electronic Texts in the Humanities S. Hockey. Design Principles for Electronic Textual Resources: Investigating Users and Uses of Scholarly Information N.J. Belkin. Section 4: Topics, Methods and Formalisms in Syntax, Semantics and Pragmatics. Evaluating English Sentences in a Logical Model J. Friedman, D.B. Moran, D.S. Warren. Recovering Implicit Information M.S. Palmer, D.A. Dahl, R.J. Schiffman, L. Hirschman, M. Linebarger, J. Downing. Flexible Generation: Taking the User into Account C.L. Paris, V.O. Mittal. Two Principles of Parse Preference J.R. Hobbs, J. Bear. UD, yet Another Unification Device R. Johnson, M. Rosner. Varieties of Heuristics in Sentence Parsing M. Nagao. Some Recent Trends in Natural Language Processing A.K. Joshi. Stone Soup and the French Room Y. Wilks.

The mitre syntactic analysis procedure for transformational grammars

A solution to the analysis problem for a class of grammars appropriate to the description of natural languages is essential to any system which involves the automatic processing of natural language inputs for purposes of man-machine communication, translation, information retrieval, or data processing. The analysis procedure for transformational grammars described in this paper was developed to explore the feasibility of using ordinary English as a computer control language.

Natural language access to structured text

This paper discusses the problem of providing natural language access to textual material. We are developing a system that relates a request in English to specific passages in a document on the basis of correspondences between the logical representations of the information in the request and in the passages. In addition, we are developing procedures for automatically generating logical representations of text passages, directly from the text, by means of an analysis of the coherence structure of the passages.

Speech understanding

Research on speech understanding is adding new dimensions to the analysis of speech and to the understanding of language. The accoustic, phonetic, and phonological processing of speech recognition efforts are being blended with the syntax, semantics, and pragmatics of question-answering systems. The goal is the development of capabilities that will allow a person to have a conversation with a computer in the performance of a shared task. Achievement of this goal will both require and contribute to a more comprehensive and powerful model of language---with significant consequences for linguistics, for computer science, and especially for computational linguistics.

The polytext system —a new design for a text retrieval system

During 1979, after preliminary studies in 1978, Kval, Stockholm, in cooperation with SRI International, California, initiated a new research and development project, planned to produce new insights into information science and to yield a useful computerized system for text retrieval, The Polytext System’s. The project is closely associated with previous and ongoing studies at the two institutions, jointly and independently, on the semantics and pragmatics of question-answering.

Computational linguistic techniques in an on-line system for textual analysis

For several years we have been involved in the development of an on-line text-processing system intended for use by information analysts in the establishment and manipulation of their own personal files. In the initial implementation of the system, a transformational syntactic analysis was applied to sentences formulated by the analyst as summaries of information content in the text he was scanning on a display scope. This analysis procedure begins with a morphological analysis and a lexical lookup that provides syntactic feature information; then, a context-free parsing takes place; finally, transformations are applied to reject inappropriate parsings, derive the base structure of the sentence, and convert the result into a canonical tree format. These canonical representations are stored and can be searched by analyzing questions in the same manner and then matching their structures against those in the data base for correspondences. On-line interaction allows the analyst to advise the program in case of ambiguity, to expand the lexicon, and to modify previous actions taken.Recently, we have added other, less sophisticated techniques to provide a range of on-line processing capabilities from simple to complex. The analyst can identify or annotate lines, paragraphs, or whole selections, creating files and subfiles for temporary or long-term storage. Text-searching procedures allow him to match on stems, words, or phrases and on combinations of any of these elements--with synonym substitutions. Sets of synonyms can be established on-line to provide a personalized thesaurus. The various components of the system can be used flexibly, and additional features can be added easily without disrupting the ones already established.The augmented system is being used now in studies which will provide information on the differential utility of these techniques in relation to the tasks that text-oriented information analysts undertake.

Natural-language-access systems and the organization and use of information

This paper describes a program of research whose objectives are to (1) develop systems that provide users with access to both data and text files through natural language dialogues; (2) study how people actually use the information to test hypotheses and solve problems; (3) modify the system designs on the basis of the results of the studies so that the systems more effectively support such uses and increasingly come to model the behavior of the users. Two of the systems are in the medical domain: the first provides physicians with formatted information derived from patient medical records; the second responds to requests by eliciting relevant passages from a medical monograph. The third system is a more general information retrieval facility that will support interactions among system users and enable their successive experiences to be accumulated within the system database.

Machine-readable dictionaries

The papers in this panel consider machine-readable dictionaries from several perspectives: research in computational linguistics and computational lexicology, the development of tools for improving accessibility, the design of lexical reference systems for educational purposes, and applications of machine-readable dictionaries in information science contexts. As background and by way of introduction, a description is provided of a workshop on machine-readable dictionaries that was held at SRI International in April 1983.

REFLECTIONS ON 20 YEARS OF THE ACL AN INTRODUCTION

Our society was founded on 13 June 1962 as the Association for Machine Translation and Computational Linguistics. Consequently, this 1982 Annual Meeting represents our 20th anniversary. We did, Of course, change our name to the Association for Computational Linguistics in 1968, but that did not affect the continuity of the organization. The date of this panel, 17 June, misses the real anniversary by four days, but no matter; the occasion still allows us to reflect on where we have been and where we are going. I seem to be sensitive to opportunities for celebrations. In looking through my AMTCL/ACL correspondence over the years, I came across a copy of a memo sent to Bob Simmons and Hood Roberts during our lOth anniversary year, recommending that something in commemoration might be appropriate. I cannot identify anything in the program of that meeting or in my notes about it that suggests they took me seriously then, but that reflects the critical difference between volunteering a recommendation and Just plain volunteerlngl