Abstract
dc:description.abstractThe problem of presenting information surrounds the need to make information relevant, personalised and engaging. As a prevalent form of presenting information narratives can be used to make information engaging and by their nature are relevant to their audience. Were we able to automaticly generate narratives for presenting information we could also personalise it, creating custom personalised narratives that selected relevant information to present and did so through an engaging experience for the audience. However existing narrative systems can either fail to generate quality narratives or sacrifice their autonomy to do so. This document presents a machine understandable thematic model for expressing themes in narratives. This model can be used in a thematic system to give a thematic subtext to the presentation of information, enriching generated narratives, and improving the thematic relevance of information. A prototype using this model in an experiment with photo montages as simple narratives demonstrated that use of the model can successfully cause results to connote desired themes and improve their relevance to titles with thematic content over simpler keyword methods. There are a variety of ways such a system could be integrated with modern narrative systems in order to enrich their results without sacrificing autonomy and the development of a thematic presenter with thematic analysis could be used to elaborate thematic content in a narrative improving its thematic cohesion and relevance.
Degree
thesis:*- Name dc:type.qualificationname
- Ph.D.
- Level dc:type.qualificationlevel
- doctoral
- Grantor dc:publisher.institution
- University of Southampton
- Year dc:date.issued
- 2009
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Hargood, Charlie
Rights
- Language dc:language.dcterms_iso639_2
- eng