MUMBLE: A Flexible System for Language Production
Proceedings of the Seventh International Joint Conference on Artificial Intelligence. Program Demonstrations. Pages 1062--1.
© 1981 International Joint Conferences on Artificial Intelligence
Abstract
- McDonald, D.D. ( 1980) Language Production as a Process of Decision-making Under Constraints, Ph.D Dissertation, MIT. (To be published as Technical Report from the MIT Artificial Intelligence Laboratory).
References were extracted automatically from the PDF and may contain errors.
Full text extracted automatically from the PDF; figures, tables and equations may differ. The PDF is the version of record.
"MUMBLE" A Flexible System for Language Production
David D. McDonald
Department of Computer and Information Science University of Massachusetts
MUMBLE is a general purpose subsystem intended as the final output segment of the natural language interrace of an expert system. As described in [ 1 ], it takes plans expressed in nearly any conceptual representation and realizes them as English texts, integrating the different parts of the plan linguistically as well as taking full responsibiUty for the grammaticality of the output
and its coherency as a discourse.
MUMBLE has been under development since 1975, having undergone four major overhauls, and is being used at MIT, BBN, and UMass. It is available from the author in a common dialect of maclisp and lispmachlne lisp, with a translation to franz lisp underway and one to interlisp in planning. The kernal of MUMBLE consists of roughly 400 functions In eleven files, and as interpreted code occupies roughly 150k words of memory on a PDP/1 0. Runtimes are a function of the complexity or the input plans, but fall around .5 to Z seconds per word or output text (again running interpreted).
Input plans for MUMBLE are invariably composed from the same data structures that the expert program uses itself to represent what it wants to talk about. They are interpreted for production by a "dictionary" that one writes for each new expert program. This dictionary associates expressions from the expert with English phrases (or phrase schemas if the expression takes arguments); should more than one phrase be possible, one either includes a set or discriminating tests or makes use or the system supplied routine for managing synonyms. Example dictionaries are available for five different domains and their internal representations.
MUMBLE supplies an extensive vocabulary of linguistic relations and operations in terms of which to write the needed English phrases, and at runtime maintains a thoroughly annotated abstract description or the linguistic structure already built from those elements of the plan that have been realized so far . This structure wfll automatically constrain the realization of the rest or the elements to insure their grammaUcallty with respect to what has gone before; it also provides a context-sensitive reference for any heuristics one may write (or adapt from those already included) to govern decisions such as the proper length of sentences, how to express coritrasts, or how elaborate a description to use for some object.
The paragraph below is an example or MUMBLE's output, starting from the given list of assertions. (The task is to readout the contents of a semantic net describing a grammar.)
((subconcept pp phrase) (has-role pp pobJ) (has·role pp prep) (has-role pp lnterp) (has-role pp ppobJ) (reatrlctlon pobJ np) (restriction prep prep) (restriction lnterp relation) (reatrlc;llon ppobJ pp) (aubconcept orpenonpp pp) (aubc;oncept lnsubjeclpp pp) (aubconcept locatlonpp pp) (aubc;oncept aboutsubjectpp pp))
Pp is the subconcept of phrase. It has the roles: pobj, prep, interp, and ppobj. Pobj must be an np, prep a prep, interp a relation, and ppobj a pp. Pp's subconcepts are ofpersonpp, insubjectpp, locationp and aboutsubjectpp.
MUMBLE's existing grammar knows how to apply rhetorical rules for discourse focus, pronominalization, ellipsis, and the merging of propositions via adjunction, relative clauses, reduced conjunctions, or apposatives. It also can plan syntactic adjustments for various thematic transformations, complex adverbials and verb groups, and all kinds of WH-movement. More importantly perhaps, MUMBLE provides a framework in which interested researchers can develop their own analyses or. the rules of language use, relieved or the need to simultaniously develop rules and control structure for all of the mundane but necessary language phenomena that go into production.
@inproceedings{ijcai1981p209,
title = {MUMBLE: A Flexible System for Language Production},
author = {McDonald, David B.},
booktitle = {Proceedings of the Seventh International Joint Conference on
Artificial Intelligence, {IJCAI-81}},
publisher = {International Joint Conferences on Artificial Intelligence},
editor = {Patrick J. Hayes},
pages = {1062---1},
year = {1981},
month = {8},
note = {Program Demonstrations},
url = {https://ijcai.org/proceedings/1981/209},
}
TY - CPAPER TI - MUMBLE: A Flexible System for Language Production AU - McDonald, David B. T2 - Proceedings of the Seventh International Joint Conference on Artificial Intelligence, IJCAI-81 PB - International Joint Conferences on Artificial Intelligence PY - 1981 SP - 1062 EP - -1 UR - https://ijcai.org/proceedings/1981/209 ER -Copied
Machine-readable versions of this paper, for tools and AI agents:
Welcome
Future Conferences
AI Journal Division
IJCAI Proceedings
About IJCAI
IJCAI Awards