Bruce W. Ballard

An experimental study of natural language programming

Abstract An experiment is described which gives data related to the usefulness and efficiency of English as a programming language. The experiment was performed with the NLC system, described herein, and used twenty-three paid volunteers from a first course in programming. Subjects were asked to solve two problems, namely (1) solution of linear equations and (2) gradebook averaging. A total of 1581 English sentences were typed, 81% of which were processed correctly. The remaining 19% were rejected because of questionable user syntax or system inadequacies. In most cases, subjects were able to paraphrase a rejected input in terminology understandable by the system. The overall success rate at solving an entire problem, within the 2½ h time constraint of the experiment, was 73·9%. In short, the system was an effective problem solver for the selected classes of problems and users. Many system failures resulted from “bugs” or syntactic oversights which appear amenable to easy repair. None of the Standard concerns about natural language programming related to vagueness, ambiguity, verbosity or correctness was a significant problem, although minor difficulties did arise occasionally.

LDC-1: a transportable, knowledge-based natural language processor for office environments

During the 1970s, a number of systems providing limited English-language processing capabilities were developed to permit computer access by casual or untrained users. The authors interest is in adapting and extending techniques developed for these systems, especially those used in database query systems and their own English-language programming language system (NLC), for use in office environments. This paper describes the layered domain class system LDC, a state-of-the-art natural language processor whose major goals are (1) to provide english-language retrieval capabilities for medium-sized office domains that have been stored on the computer as text-edited files, as opposed to more restrictive database structure; and (2) to eliminate the need to call in the system designer when extensions into new domains are desired, without sacrificing the depth or reliability of the interface. In this paper the authors (a) provide an overview of LDC, including sample inputs; (b) briefly discuss the role of each module of the system, with special attention to provisions for users to adapt the system to deal with new types of data; and (c) consider the relation of our system to other formal and natural language interfaces that are in use or under development. 77 references.

Programming in natural language: “NLC” as a prototype

The state of the art in computational linguistics has progressed to the point where it is now possible to process simple programs written in natural language. This report describes a natural language programming system called NLC which enables a computer user to type English commands into a display terminal and watch them executed on example data shown on the screen. The system is designed to process data stored in matrices or tables, and any problem which can be represented in such structures can be handled if the total storage requirements are not excessive.

INTERACTIVE NATURAL LANGUAGE PROBLEM SOLVING: A PRAGMATIC APPROACH

A class of natural language processors is described which allow a user to display objects of interest on a computer terminal and manipulate them via typed or spoken English sentences.This paper concerns itself with the implementation of the voice input facility using an automatic speech recognizer, and the touch input facility using a touch sensitive screen. To overcome the high error rates of the speech recognizer under conditions of actual problem solving in natural language, error correction software has been designed and is described here. Also described are problems involving the resolution of voice input with touch input, and the identification of the intended referents of touch input.To measure system performance we have considered two classes of factors: the various conditions of testing, and the level and quality of training of the system user. In the paper a sequence of five different testing situations is observed, each one resulting in a lowering of system performance by several percentage points below the previous one. A training procedure for potential users is described, and an experiment is discussed which utilizes the training procedure to enable users to solve actual non-trivial problems using natural language voice communication.

THE SYNTAX AND SEMANTICS OF USER-DEFINED MODIFIERS IN TRANSPORTABLE NATURAL LANGUAGE PROCESSOR

The Layered Domain Class system (LDC) is an experimental natural language processor being developed at Duke University which reached the prototype stage in May of 1983. Its primary goals are (1) to provide English-language retrieval capabilities for structured but unnormalized data files created by the user; (2) to allow very complex semantics, in terms of the information directly available from the physical data file; and (3) to enable users to customize the system to operate with new types of data. In this paper we shall discuss (a) the types of modifiers LDC provides for; (b) how information about the syntax and semantics of modifiers is obtained from users; and (c) how this information is used to process English inputs.

Transportable English-language processing for office environments

This article describes the Layered Domain Class system (LDC), a state-of-the-art natural language processor whose major goals are (1) to provide English-language retrieval capabilities for medium-sized office domains that have been stored on the computer as text-edited files, rather than more restrictive database structures and (2) to eliminate the need to call in the system designer when extensions into new domains are desired, without sacrificing the depth or reliability of the interface. Early developments in the design of portions of LDC were presented at NCC-83, and the entire system became operational in July 1983. The article gives an overview of the construction of the system, gives examples of the English structures provided for, briefly describes the most recently completed portions of the system, and mentions current directions the project is taking.

Natural Language programming: Duke University

This project concentrates on the fast, efficient execution of sequences of imperative sentences for a variety of applications. The general strategy assumes that the machine controls a world of interest to the user which can be observed on a computer display screen. The user inputs sentences requesting actions to be performed on the world and observes the result of each command. If the result does not satisfy the user, he or she may request a back-up and rephase the command in clearer language. Most commands are executed in about 2 to 4 seconds making it possible to have a genuine dialog in real time in the solution of real problems.

A system for natural language computation

This paper briefly describes a natural language programming system called NLC which allows users to type commands in English and watch the resulting actions carried out on a computer display screen. The data entities appear on the screen at all times, so that the user can observe the effect of each input before proceeding to the next one. Should some input ever be incorrectly processed, a backup facility allows requests to be clarified after restoring the previous context and data values. The user may identify a sequence of natural language commands as a procedure which, when given a name, acts as a new imperative verb and may be used to define further customizations.