2026-09-19T16:21:46Zhttps://dspace.mit.edu/server/oai/requestoai:dspace.mit.edu:1721.1/664402022-01-13T07:54:29Zcom_1721.1_7582com_1721.1_7581com_1721.1_104841col_1721.1_131023col_1721.1_141484
Patrick Winston.
Low, Harold William Capen, IV
Massachusetts Institute of Technology. Dept. of Electrical Engineering and Computer Science.
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
2011-10-17T21:26:16Z
2011-10-17T21:26:16Z
2011
2011
http://hdl.handle.net/1721.1/66440
755631697
Thesis (M. Eng.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2011.
Cataloged from PDF version of thesis.
Includes bibliographical references (p. 72).
Whether through anecdotes, folklore, or formal history, humans learn the lessons and expectations of life from stories. If we are to build intelligent programs that learn as humans do, such programs must understand stories as well. Casting narrative text in an information-rich representation affords Al research platforms, such as the Genesis system, the capacity to understand the events of stories individually. To understand a story, however, a program must understand not just events, but also how events cause and motivate one another. In order to understand the relationships between these events, stories must be saturated with implicit details, connecting given events into coherent plot arcs. In my research, my first step was to analyze a range of story summaries in detail. Using nearly 50 rules, applicable to brief summaries of stories taken from international politics, group dynamics, and basic human emotion, I demonstrate how a rendition of Frank Herbert's Dune can be automatically understood so as to produce an interconnected story network of over one hundred events. My second step was to explore the nuances of rule construction, finding which rules are needed to create story networks reflective of proper implicit understanding and how we, as architects, must shape those rules to be understood. In particular, I develop a method that constructs new rules using the rules already embedded in stories, a representation of higher-order thinking that enables us to speak of our ideas as objects.
by Harold William Capen Low, IV.
M.Eng.
72 p.
eng
Massachusetts Institute of Technology
M.I.T. theses are protected by
copyright. They may be viewed from this source for any purpose, but
reproduction or distribution in any format is prohibited without written
permission. See provided URL for inquiries about permission.
http://dspace.mit.edu/handle/1721.1/7582
Electrical Engineering and Computer Science.
Story understanding in Genesis : exploring automatic plot construction through commonsense reasoning
Thesis
application/pdf
false
Publication
unknown
unknown
<Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="81369d32-628c-4bc3-a8dc-549fc1f5e8c6">
<Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type>
<Language>eng</Language>
<Title>Story understanding in Genesis : exploring automatic plot construction through commonsense reasoning</Title>
<PublishedIn>
<Publication>
</Publication>
</PublishedIn>
<PublicationDate>2011</PublicationDate>
<Authors>
<Author>
<DisplayName>Low, Harold William Capen, IV</DisplayName>
<Affiliation>
<OrgUnit>
</OrgUnit>
</Affiliation>
</Author>
</Authors>
<Editors>
</Editors>
<Publishers>
<Publisher>
<DisplayName>Massachusetts Institute of Technology</DisplayName>
<OrgUnit />
</Publisher>
</Publishers>
<License>http://dspace.mit.edu/handle/1721.1/7582</License>
<Keyword>Electrical Engineering and Computer Science.</Keyword>
<Abstract>Whether through anecdotes, folklore, or formal history, humans learn the lessons and expectations of life from stories. If we are to build intelligent programs that learn as humans do, such programs must understand stories as well. Casting narrative text in an information-rich representation affords Al research platforms, such as the Genesis system, the capacity to understand the events of stories individually. To understand a story, however, a program must understand not just events, but also how events cause and motivate one another. In order to understand the relationships between these events, stories must be saturated with implicit details, connecting given events into coherent plot arcs. In my research, my first step was to analyze a range of story summaries in detail. Using nearly 50 rules, applicable to brief summaries of stories taken from international politics, group dynamics, and basic human emotion, I demonstrate how a rendition of Frank Herbert's Dune can be automatically understood so as to produce an interconnected story network of over one hundred events. My second step was to explore the nuances of rule construction, finding which rules are needed to create story networks reflective of proper implicit understanding and how we, as architects, must shape those rules to be understood. In particular, I develop a method that constructs new rules using the rules already embedded in stories, a representation of higher-order thinking that enables us to speak of our ideas as objects.</Abstract>
<Access xmlns="http://purl.org/coar/access_right"
>
</Access>
</Publication>