Computational Generation of Referring Expressions Containing Relational Descriptions
| dc.contributor.advisor | Sanders, Nathan | |
| dc.contributor.author | Abernathy, Celia | |
| dc.date.accessioned | 2013-06-07T17:49:09Z | |
| dc.date.available | 2013-06-07T17:49:09Z | |
| dc.date.issued | 2013 | |
| dc.description.abstract | Referring expressions like the tall red cup are used to identify entities in many different language tasks, so referring expression generation (REG) is a crucial part of computational natural language generation. One of the challenges of REG is relational descriptions that refer to an object by relating it to some other object, as in the cup on the table. This paper presents a new computational algorithm, DR-REG, for referring expression generation using relational descriptions. DRREG generates relational referring expressions for complex objects by searching through syntactic and semantic information for a suitable expression. DR-REG is evaluated in a controlled experiment that shows its referring expressions lead readers to correctly identify a target object. | en |
| dc.description.sponsorship | Swarthmore College. Dept. of Linguistics | en |
| dc.identifier.uri | http://hdl.handle.net/10066/10895 | |
| dc.language.iso | en_US | en |
| dc.rights | Full copyright to this work is retained by the student author. It may only be used for non-commercial, research, and educational purposes. All other uses are restricted. | |
| dc.title | Computational Generation of Referring Expressions Containing Relational Descriptions | en |
| dc.type | Thesis (B.A.) |