Domain Independent Ordering of Narrative Events

Authors

Lucian T. Silcox, University of Texas-Pan American

Date of Award

12-2012

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Computer Science

First Advisor

Dr. Emmett Tomai

Second Advisor

Dr. Richard Fowler

Third Advisor

Dr. Laura Grabowski

Abstract

A key aspect towards gaining a rich understanding of a text is the ability to order the events that occur in the narrative. For us, as readers, this is a simple task, accomplished without effort by recognizing subtle contextual and linguistic clues. For natural language systems, however, the task is considerably more difficult. It is the intent of this research to automate the process of annotating and ordering temporal events in a narrative, and to measure the success of techniques trained on news articles when applied to less sequentially-structured texts. We build on the earlier work of Mani et al (2005) and Chambers et al (2007), applying machine learning techniques to the Timebank Corpus, and further, to a biographical corpus compiled for the purposes of this study.

Comments

Copyright 2012 Lucian Silcox. All Rights Reserved.

https://www.proquest.com/dissertations-theses/domain-independent-ordering-narrative-events/docview/1530412859/se-2

Granting Institution

University of Texas-Pan American