Norman K. Sondheimer

Spatial reference and natural-language machine control

Current research on natural-language speech-understanding provides encouragement for the development of systems for the vocal control of mechanical devices. However, the designer of such systems faces a variety of difficulties in allowing for references to the position, orientation, and direction of motion of objects and actions in space. This paper analyzes sources of these difficulties and conceivable solutions to them.

An Improved Heuristic for Ellipsis Processing

Several natural language systems (e.g., Bobrow et al., 1977; Hendrix et al., 1978; Kwasny and Sondheimer, 1979) include heuristics for replacement and repetition ellipsis, but not expansion ellipsis. One general strategy has been to substitute fragments into the analysis of the previous input, e.g., substituting parse trees of the elliptical input into the parse trees of the previous input in LIFER (Hendrix, et al., 1978). This only applies to inputs of the same type, e.g., repeated questions.

A semantic analysis of reference to spatial properties

A uniform analysis is offered for the source of the locations specified by all references in English to spatial properties including location and movement. This source is argued to be the location of events and states of affairs. These locations are specified by sets showing spaces momentarily occupied. Descriptions of motion are accounted for through a variety of ways of referencing these sets. Some classes of simple clauses are identified as requiring semantic analysis involving multiple events and states of affairs. This importantly allows spatial references to be associated with different events and states of affairs. Identification of object location is accounted for by use of extra inference rules or meaning postulates. A number of other explanations are suggested for more limited and derivative phenomena. The analysis is developed within a Case-like notation presented in the first-order predicate calculus. It is supported by an extensive analysis of spatial reference phenomena and by identification of ancillary benefits.

An experimental graphics system with natural language input

Abstract An experimental graphics system is described which allows interaction through natural language and touches on the screen. Artificial Intelligence techniques for processing language and representing knowledge are investigated as adequate tools with which to construct a habitable graphics system. The system is evaluated and conclusions presented.

EVALUATION OF NATURAL LANGUAGE INTERFACES TO DATABASE SYSTEMS: A PANEL DISCUSSION

For a natural language access to database system to be practical it must achieve a good match between the capabilities of the user and the requirements of the task. The user brings his own natural language and his own style of interaction to the system. The task brings the questions that must be answered and the database domaln+s semantics. All natural language access systems achieve some degree of success. But to make progress as a field, we need to be able to evaluate the degree of this success.

ON THE PRESENT

There are certainly many reasons. One is the attractiveness of our most abstract theories. They are widely presented and receive the most scholarly attention. The popular and technical press contributes by pub1~cizing our w~]der claims and broadest hopes. ~imilar]y, the press oversells our current systems, leading more careful observers to wonder even about these. Finally, mechanizing the understanding of natural language ~s very difficult. We can not hope to achieve many of our goals in the near future. Making do with the technology now available is very frustrating. All this contributes to we the members of the field gravitating to theorizing and small laboratory studies. We are choosing to focus on the £uture rather than the present. There is a real danger in this state of affairs. The build up of public and institutional expectations without a corresponding emergence of useful systems will produce a counter reaction. We have seen it before. To this day, machine translation research in the United States has not completely recovered. There is more need than ever~ there is more technology than before, word processing and computer typesetting have changed the price equation, but it is stilS not considered wise to be associated with MT. We can not let this sort of reversal happen to us again. Fortunately, we need not.

Natural language processing systems and III-formed input: University of Delaware/Sperry Univac

The goal of our project is to add robustness to natural language understanding systems by adding rule-based methods of handling illformed input [1,2,3]. This includes both ungrammatical and semantically inappropriate utterances. The method being used consists of the following procedure:

Natural language graphics

The Natural Language Graphics Project (NLG) at The Ohio State University is concerned with natural language programming in an interactive graphics environment. Natural interaction between men depends on a combination of both graphical and linguistic modes of communication [1]. The goals of the project are to study the relationships between graphical and linguistic processing in a programming environment. To achieve these goals, associated practical and theoretical issues in man-machine interaction, computer graphics, computational linguistics, and knowledge representation will be investigated.