• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Sprache und Datenverarbeitung >Plans and Planning in Narrative Generation: A Review of Plan-Based Approaches to the Generation of Story, Discourse and Interactivity in Narratives
【24h】

Plans and Planning in Narrative Generation: A Review of Plan-Based Approaches to the Generation of Story, Discourse and Interactivity in Narratives

机译：叙事生成中的计划和规划：叙事中故事，话语和互动生成的基于计划的方法的回顾

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

A range of work has made use of plan-based knowledge representations and reasoning approaches to produce elements of narrative in both interactive and non-interactive environments. The close alignment between existing AI plan representations and both narrative theoretic and cognitive models of narrative structure has greatly facilitated work on the generation of story, discourse and interactivity. Narrative theoretic and cognitive/ comprehension-focused models of narrative contain a range of elements that address issues beyond those paralleled by AI plan representations. For example, narratological concepts like unreliable narrators or focalization (Bal, 1997) are not specifically dealt with by planning models. Similarly, cognitive psychologists' models of salience in narrative understanding (Zwaan & Radvansky, 1998) are not found in plan generation approaches. Most AI approaches to narrative generation that leverage planning models use them as a base or a foundation, viewing them as the primitive buildings blocks that make up narrative structure. Additional work can then build on the primitive model to characterize more complex correlates. Examples of this type of approach include Bae and his collaborators' (2011) plan-based characterization of focalization and Cardone-Rivera and his colleagues' (2012) definition of Indexter, a plan-based model of salience in the context of story understanding.
机译：一系列工作已经利用基于计划的知识表示和推理方法在交互式和非交互式环境中产生叙述元素。现有的AI计划表示与叙事结构的叙事理论和认知模型之间的紧密结合，极大地促进了故事，话语和互动性的产生。叙事理论和以认知/理解为重点的叙事模型包含一系列解决AI计划表示形式无法解决的问题的要素。例如，叙事学概念，如不可靠的叙事者或焦点人物（Bal，1997）并未得到计划模型的专门处理。同样，认知心理学家在叙事理解中的显着性模型（Zwaan＆Radvansky，1998）在计划生成方法中也没有发现。大多数利用计划模型进行叙事生成的AI方法都将它们作为基础或基础，将它们视为构成叙事结构的原始构建块。然后，可以在原始模型上进行其他工作来表征更复杂的相关性。这种类型的方法的例子包括Bae和他的合作者（2011年）基于计划的集中化特征，以及Cardone-Rivera和他的同事（2012年）对Indexter的定义，Indexter是故事理解中基于计划的显着性模型。

著录项

  • 来源
    《Sprache und Datenverarbeitung》 |2013年第2期|41-64193-194|共26页
  • 作者

    R. Michael Young; Stephen Ware; Brad Cassell; Justus Robertson;

  • 作者单位

    North Carolina State University Department of Computer Science Campus Box 8206, Raleigh NC 27695-8206, USA;

    University of New Orleans Computer Science Department Math Building, 3rd Floor, Room 337 New Orleans LA, 70148, USA;

    North Carolina State University Department of Computer Science Campus Box 8206, Raleigh, NC 27695-8206, USA;

    North Carolina State University Department of Computer Science Campus Box 8206, Raleigh, NC 27695-8206, USA;

  • 收录信息
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号