國立虎尾科技大學 |

A Computational Model for The Portrayal of Personality Traits in Planning-Based Narrative Generation.

紀錄類型:	書目-語言資料,手稿 : Monograph/item
正題名/作者:	A Computational Model for The Portrayal of Personality Traits in Planning-Based Narrative Generation./
作者:	Bahamon, Julio Cesar.
面頁冊數:	1 online resource (252 pages)
附註:	Source: Dissertation Abstracts International, Volume: 78-08(E), Section: B.
標題:	Computer science. -
電子資源:	click for full text (PQDT)
ISBN:	9781369620078

A Computational Model for The Portrayal of Personality Traits in Planning-Based Narrative Generation.
Bahamon, Julio Cesar.

A Computational Model for The Portrayal of Personality Traits in Planning-Based Narrative Generation. - 1 online resource (252 pages)

Source: Dissertation Abstracts International, Volume: 78-08(E), Section: B.

Thesis (Ph.D.)--North Carolina State University, 2016.

Includes bibliographical references

Storytelling is an essential component of human culture. Humans use narratives to describe, understand, and relate to events of the world in which they live. Narrative plays a key role as a means to transfer knowledge. This is one of the principal motivations for Artificial Intelligence research that centers around the development of techniques and architectures to automate the generation of narrative content. Computer-generated narratives can be applied in multiple domains, such as training simulations, activity visualizations, instructional videos, and digital games.

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

Mode of access: World Wide Web

ISBN: 9781369620078Subjects--Topical Terms:

573171
Computer science.
Index Terms--Genre/Form:

554714
Electronic books.

A Computational Model for The Portrayal of Personality Traits in Planning-Based Narrative Generation.
LDR:06001ntm a2200361K 4500 001 913685
005 20180622095235.5
006 m o u
007 cr mn||||a|a||
008 190606s2016 xx obm 000 0 eng d
020 $a 9781369620078
035 $a (MiAaPQ)AAI10583312
035 $a AAI10583312
040 $a MiAaPQ $b eng $c MiAaPQ
100 1 $a Bahamon, Julio Cesar. $3 1186621
245 1 2 $a A Computational Model for The Portrayal of Personality Traits in Planning-Based Narrative Generation.
264 0 $c 2016
300 $a 1 online resource (252 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: 78-08(E), Section: B.
500 $a Adviser: R. Michael Young.
502 $a Thesis (Ph.D.)--North Carolina State University, 2016.
504 $a Includes bibliographical references
520 $a Storytelling is an essential component of human culture. Humans use narratives to describe, understand, and relate to events of the world in which they live. Narrative plays a key role as a means to transfer knowledge. This is one of the principal motivations for Artificial Intelligence research that centers around the development of techniques and architectures to automate the generation of narrative content. Computer-generated narratives can be applied in multiple domains, such as training simulations, activity visualizations, instructional videos, and digital games.
520 $a One of the key elements that can improve the effectiveness of a story is the presence of interesting and compelling characters. Well-developed characters have features that enable them to significantly enhance the believability and quality of a narrative. In my research, I focus on the development of a computational model aimed at facilitating the inclusion of compelling characters in narrative that is automatically generated by a planning-based system. The model centers on the use of an intelligent process to express character personality. In this model, personality is operationalized as behavior that results from choices made by a character in the course of a story. This operationalization uses the taxonomy defined in the Five Factor Model (FFM) and results from behavioral psychology studies that link behavior to personality traits. I hypothesize that the relationship between choices and the actions they lead to can be used in narrative to produce the perception of specific personality traits in an audience.
520 $a The design of my narrative generation model is also influenced by the theory of human affect proposed by Lazarus. This theory is of particular relevance because I use it to simulate the mental process that a fictional character goes through as she deliberates on a course of action to follow. Affect theory informs the mechanism that is used to compare and contrast multiple courses of action that achieve a specific character intention, but that could have either beneficial or adverse effects for other characters in the story. The mechanism quantifies effects as being adverse or beneficial in terms of whether they help or prevent characters from achieving their intentions (things they want to accomplish). An example of a character intention can be to obtain wealth. The mechanism that I designed informs the work of the narrative generation algorithm to favor the creation of stories with actions and action sequences that highlight character personality traits. To achieve this, the mechanism evaluates characters' actions within the context of their intentions and those of other characters in the story.
520 $a The system that I designed, named Mask, has its foundation on the use of state-space planning to generate narratives where characters act intentionally. My implementation uses the Glaive narrative planner that was developed by Ware and Young. The version I use has been significantly modified and enhanced to include the functionality described above. In contrast to a traditional narrative planner where a uniform planning process handles the construction of a story, in choice-based narrative generation, planning is interleaved between a macro-planner and a micro-planner. The macro-planner is responsible for constructing the authorial view of the global story and its causal coherence. The micro-planner is a restricted version of a planning system, designed to simulate the character's decision-making process as she works toward accomplishing a goal.
520 $a I conducted an initial evaluation with human subjects using handcrafted stories that were created following the algorithm I designed. Results from this evaluation were encouraging yet inconclusive. As a result, I conducted a formative study after revising my protocol and in particular the discourse method used to present the stories to the audience. Results from this study were significant enough to reject the null hypothesis and support the claim that an audience presented with a set of alternate action sequences for a story character will select personality trait ratings that have significant correlation with the choices made by such character. Additionally, the data collected from the formative study informed the fine-tuning and implementation of my narrative generation model. An evaluation of the completed Mask narrative generation system was conducted using automatically generated stories that were produced in accordance with the principles described above. Results from this evaluation were comparable to those from the formative study, i.e., the computer-generated stories elicited the perception of specific character personality traits on the audience that experienced such stories.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2018
538 $a Mode of access: World Wide Web
650 4 $a Computer science. $3 573171
650 4 $a Artificial intelligence. $3 559380
655 7 $a Electronic books. $2 local $3 554714
690 $a 0984
690 $a 0800
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a North Carolina State University. $b Computer Science. $3 1182345
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10583312 $z click for full text (PQDT)