Narrative Complexity by Unknown

Author:Unknown
Language: eng
Format: epub
Tags: LIT006000 Literary Criticism / Semiotics & Theory, SOC052000 Social Science / Media Studies
Publisher: University of Nebraska Press

Microworlds: Complexity Reduction through Domain Specificity

Limiting language to specific domains, which presumably involve a closed subset of a given natural language, is one of the key avenues AI pursued in language research. The following are early examples of this approach.

STUDENT was a program developed to solve algebraic story problems such as this: if the number of customers Tom gets is twice the square of 20 percent of the number of advertisements he runs, and the number of advertisements he runs is 45, what is the number of customers Tom gets? (Bobrow 1964).To solve this problem, STUDENT looked for cues such as “number,” “twice,” “square,” and so on to break the sentence into chunks, assigned variables to these chunks, formed equations, and solved them. A typical stumbling block of its approach to language was that it interpreted the phrase “the number of times I went to the movies” as the product of the two variables “the number of” and “I went to the movies” (Russell and Norvig 1995, 11).

SHRDLU was a simulation of a robotic hand that moved colored blocks around in a virtual world shown on a monitor, following instructions given to it in natural language. It could precisely follow instructions as complicated as “Find a block which is taller than the one you are holding and put it into the box” or “Move the red cube to the left of the big block to the box” (Winograd 1972, 66). A simulated robot built at MIT, which operated in a simulated-blocks microworld, SHRDLU was equipped with a graphic interface and a video screen that displayed its operations in a visual manner. Furthermore, it had a written-language interface through which it not only followed commands given in ordinary English but also answered questions about its “motivations” for doing things in a certain sequence. When asked why it picked up the little pyramid among its blocks, for instance, it replied, “to clear off the red cube.” Moreover, it was also able to remember and report its previous actions, such as touching a particular pyramid before putting a particular block on a particular small cube (Crevier 1993). With these capabilities, SHRDLU boasted as an impressive system in its time. For all its linguistic virtuosity, though, SHRDLU was very brittle and was rigidly restricted to the blocks domain.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.