國立虎尾科技大學 |

Investigations of the Properties of Narrative Schemas.

Record Type:	Language materials, manuscript : Monograph/item
Title/Author:	Investigations of the Properties of Narrative Schemas./
Author:	Simonson, Daniel E.
Description:	1 online resource (259 pages)
Notes:	Source: Dissertation Abstracts International, Volume: 79-09(E), Section: A.
Contained By:	Dissertation Abstracts International79-09A(E).
Subject:	Linguistics. -
Online resource:	click for full text (PQDT)
ISBN:	9780355871111

Investigations of the Properties of Narrative Schemas.
Simonson, Daniel E.

Investigations of the Properties of Narrative Schemas. - 1 online resource (259 pages)

Source: Dissertation Abstracts International, Volume: 79-09(E), Section: A.

Thesis (Ph.D.)--Georgetown University, 2018.

Includes bibliographical references

Narrative schemas are generalizations of frequently re-occurring sequences of events linked through co-referring entities in text (Chambers & Jurafsky, 2009). The use of such schemas in the unsupervised analysis of text promises to assist with the characterization, comparison, and analysis of large volumes of text. However, problems exist prior to conducting such an analysis, particularly with respect to evaluation. Most work following Chambers & Jurafsky (2009) focuses on cloze task performance, which does not directly evaluate schemas. To this end, I devise techniques to directly measure properties of narrative schemas.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2018

Mode of access: World Wide Web

ISBN: 9780355871111Subjects--Topical Terms:

557829
Linguistics.
Index Terms--Genre/Form:

554714
Electronic books.

Investigations of the Properties of Narrative Schemas.
LDR:03646ntm a2200373Ki 4500 001 919346
005 20181127124955.5
006 m o u
007 cr mn||||a|a||
008 190606s2018 xx obm 000 0 eng d
020 $a 9780355871111
035 $a (MiAaPQ)AAI10808470
035 $a (MiAaPQ)georgetown:14036
035 $a AAI10808470
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Simonson, Daniel E. $3 1193896
245 1 0 $a Investigations of the Properties of Narrative Schemas.
264 0 $c 2018
300 $a 1 online resource (259 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertation Abstracts International, Volume: 79-09(E), Section: A.
500 $a Adviser: Anthony R. Davis.
502 $a Thesis (Ph.D.)--Georgetown University, 2018.
504 $a Includes bibliographical references
520 $a Narrative schemas are generalizations of frequently re-occurring sequences of events linked through co-referring entities in text (Chambers & Jurafsky, 2009). The use of such schemas in the unsupervised analysis of text promises to assist with the characterization, comparison, and analysis of large volumes of text. However, problems exist prior to conducting such an analysis, particularly with respect to evaluation. Most work following Chambers & Jurafsky (2009) focuses on cloze task performance, which does not directly evaluate schemas. To this end, I devise techniques to directly measure properties of narrative schemas.
520 $a I first define the distinction between score---what is evaluated on the cloze task---and germinator---how a score is used to generate schemas. I re-interpret Chambers & Jurafsky (2009)'s technique for generating schemas in these terms and devise two novel schema germination techniques. These different germinators produce very different sets of schemas. To evaluate schemas directly, I create two new tasks. The first is the Narrative Argument Salience Through Entities Annotated task, where schemas are shown to generally perform better than a number of baselines. I also coin a pair of minimum description length inspired measures. I conduct a meta-evaluation of these measures on the OntoNotes corpus (Weischedel et al., 2013); they show an insignificant degradation in schema quality despite receiving higher quality data.
520 $a I also examine the relationship between schemas, document categories, and topics. The presence of specific schemas does not predict document category well; however, document categories seem to determine the schemas generated, especially in the case of homogenous categories, such as the Weddings and Obituaries, which are significantly different from their heterogeneous counterparts. Topic models generated through latent dirichlet allocation (Blei et al., 2003) do not seem to exhibit this behavior in the model proposed here; however, topics treated similarly to document categories may provide a similar benefit.
520 $a Finally, I investigate the stability of narrative schemas, generating schemas based on 100 different ablations and cross-validations of the NYT Corpus. Generally, it seems that more documents reduce the stability of schemas, likely influenced by the endless variation of the world reflected in news narratives.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2018
538 $a Mode of access: World Wide Web
650 4 $a Linguistics. $3 557829
650 4 $a Computer science. $3 573171
655 7 $a Electronic books. $2 local $3 554714
690 $a 0290
690 $a 0984
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a Georgetown University. $b Linguistics. $3 1182048
773 0 $t Dissertation Abstracts International $g 79-09A(E).
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10808470 $z click for full text (PQDT)