国产bbaaaaa片,成年美女黄网站色视频免费,成年黄大片,а天堂中文最新一区二区三区,成人精品视频一区二区三区尤物

首頁> 外文學位 >Abstraction of reasoning for problem solving and tutoring assistants .
【24h】

Abstraction of reasoning for problem solving and tutoring assistants .

機譯：解決問題的推理和輔導助手的抽象。

獲取原文
獲取原文并翻譯 | 示例
  • 摘要
  • 著錄項
  • 相似文獻
  • 相關主題

摘要

This dissertation presents an approach to the abstraction of the reasoning of a knowledge-based agent that facilitates human-agent collaboration in complex problem solving and decision-making and the development of systems for tutoring expert problem solving to non-experts.;Effective human-agent collaboration requires an ability of the user to easily understand the complex reasoning generated by the agent. The methods presented in this dissertation allow the partition of a complex reasoning tree into meaningful and manageable sub-trees, the abstraction of individual sub-trees, and the automatic generation of an abstract tree that plays the role of a table of contents for the display, understanding and navigation of the concrete tree.;Abstraction of reasoning is also very important for teaching complex problem-solving to non-experts. This dissertation presents a set of integrated methods that allow the abstraction of complex reasoning trees to define abstract problem solving strategies for tutoring, the rapid development of lesson scripts for teaching these strategies to non-experts, and the automatic generation of domain-specific lessons. These methods are augmented with ones for learning and context-sensitive generation of omission, modification, and construction test questions, to assess a student's problem solving knowledge.;The developed methods have been implemented as an extension of the Disciple learning agent shell and have led to the development of the concept of learning and tutoring agent shell. This is a general tool for building a new type of intelligent assistants that can learn complex problem solving expertise directly from human experts, support human experts in problem solving and decision making, and teach their problem solving expertise to non-experts. The developed learning and tutoring shell has been used to build a prototype tutoring system in the intelligence analysis domain which has been used and evaluated in courses at the US Army War College and George Mason University.
機譯：本文提出了一種基于知識的智能體推理的抽象方法，該方法可以促進復雜的問題解決和決策過程中的人與智能體之間的協(xié)作，以及為非專家提供專家解決問題的輔導系統(tǒng)的開發(fā)。座席協(xié)作需要用戶具有輕松理解座席生成的復雜推理的能力。本文提出的方法允許將復雜的推理樹劃分為有意義的和可管理的子樹，對單個子樹進行抽象，并自動生成充當顯示目錄的抽象樹，理解和導航具體樹。推理的抽象對于向非專家教授復雜的問題解決方法也非常重要。本論文提出了一套綜合的方法，這些方法允許抽象的復雜樹定義定義用于輔導的抽象問題解決策略，向非專家教授這些策略的課程腳本的快速開發(fā)以及領域特定課程的自動生成。這些方法增加了用于學習和上下文敏感的遺漏，修飾和構造測試問題的方法，以評估學生的解決問題的知識。；已開發(fā)的方法已實現(xiàn)為門徒學習代理外殼的擴展，并導致以發(fā)展學習和輔導代理殼的概念。這是構建新型智能助手的通用工具，該智能助手可以直接從人類專家那里學習復雜的解決問題的專業(yè)知識，為人類專家解決問題和制定決策提供支持，并向非專家教授他們的解決問題的專業(yè)知識。開發(fā)的學習和輔導殼已用于在情報分析領域構建原型輔導系統(tǒng)，該系統(tǒng)已在美國陸軍戰(zhàn)爭學院和喬治·梅森大學的課程中使用和評估。

著錄項

  • 作者

    Le, Vu.;

  • 作者單位

    George Mason University.;

  • 授予單位 George Mason University.;
  • 學科 Artificial Intelligence.
  • 學位 Ph.D.
  • 年度 2008
  • 頁碼 237 p.
  • 總頁數(shù) 237
  • 原文格式 PDF
  • 正文語種 eng
  • 中圖分類 人工智能理論;
  • 關鍵詞

相似文獻

  • 外文文獻
  • 中文文獻
  • 專利
獲取原文

客服郵箱：kefu@zhangqiaokeyan.com

京公網(wǎng)安備：11010802029741號 ICP備案號：京ICP備15016152號-6 六維聯(lián)合信息科技 (北京) 有限公司?版權所有

  • 客服微信

  • 服務號