Files in this item
|(no description provided)|
|Title:||Communication and Miscommunication (Language, Artificial Intelligence, Natural Understanding)|
|Author(s):||Goodman, Bradley Alan|
|Department / Program:||Computer Science|
|Degree Granting Institution:||University of Illinois at Urbana-Champaign|
|Abstract:||This thesis discusses one aspect of enabling people to communicate in natural language with computers. The central focus of this work is a study on how one could build robust natural language processing systems that can detect and recover from miscommunication. The study of miscommunication is a necessary task within such a context since any computer capable of communicating with humans in natural language must be tolerant of the imprecise, ill-devised or complex utterances that people often use. This goal first requires an inquiry into how people communicate and how they recover from problems in communication. That investigation centers on the kinds of miscommunication that occur in human communication with a special emphasis on reference problems, i.e., problems a listener has determining whom or what a speaker is talking about. A collection of protocols of a speaker explaining to a listener how to assemble a toy water pump were studied and the common errors seen in speakers' descriptions were categorized. This study led to the development of techniques for avoiding failures of reference that were employed in the reference identification component of a natural language understanding program.
The traditional approaches to reference identification in previous natural language systems were found to be less elaborate than people's real behavior. In particular, listener's often find the correct referent even when the speaker's description does not describe any object in the world. To model a listener's behavior, a new component was added to the traditional reference identification mechanism to resolve difficulties in a speaker's description. This new component uses knowledge about linguistic and physical context in a negotiation process that determines the most likely places for error in the speaker's utterance. The actual repair of the speaker's description is achieved by using the knowledge sources to apply relaxation techniques that delete or replace portions of the description. The algorithm developed more closely approximates people's behavior.
Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 1984.
|Date Available in IDEALS:||2014-12-15|