• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Web-based education >HELP DESK ARCHITECTURE FOR A U-LEARNING SYSTEM CHARACTERIZED BY THE USE OF PROBABILISTIC REASONING-BASED SEARCH OF A CASE BASE
【24h】

HELP DESK ARCHITECTURE FOR A U-LEARNING SYSTEM CHARACTERIZED BY THE USE OF PROBABILISTIC REASONING-BASED SEARCH OF A CASE BASE

机译：基于案例推理的案例数据库搜索表征的U学习系统的帮助平台架构

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

This paper proposes an architecture for a help desk for a u-learning system. The architecture is characterized by the use of a case base search based on probabilistic reasoning. This case base search method uses indices that represent the relationships between key words and cases stored in a case base, using a Bayesian network. The Noisy-or rule is used to simplify the setting of the conditional probability between a key word node and a case node, and also the computation needed to select candidate cases. We have developed an experimental case base that contains typical Q&As likely to occur in a u-learning system, and examined the performance of this indexing method. It is shown that the use of a history database to tune the index parameters can dramatically improve the probability of retrieving the correct cases. Finally, comparisons of the proposed method with the conventional D-HS method and with the Dialog Navigator are given.
机译：本文提出了一种用于u学习系统的服务台的体系结构。该架构的特点是使用基于概率推理的案例库搜索。这种基于案例的搜索方法通过贝叶斯网络使用索引来表示关键词与存储在案例库中的案例之间的关系。 Noisy-or规则用于简化关键字节点和案例节点之间的条件概率的设置，以及简化选择候选案例所需的计算。我们已经开发了一个实验案例库，其中包含可能在u学习系统中发生的典型问答，并研究了此索引方法的性能。结果表明，使用历史数据库来调整索引参数可以极大地提高检索正确案例的可能性。最后，比较了所提出的方法与传统的D-HS方法和Dialog Navigator。

著录项

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