• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Second ICSC Symposium on Engineering of Intelligent Systems, Jun 27-30, 2000, Scotland, U.K. >Agent IDA in Model-Based Diagnostic Domain
【24h】

Agent IDA in Model-Based Diagnostic Domain

机译：基于模型的诊断域中的Agent IDA

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

The main characteristic of a model-based diagnostic approach is the use of a deep knowledge for a diagnostic reasoning. Diagnostic result is a set of declarations about components' behavior called diagnosis, which could explain unexpected system behavior. Generally, there may be several competing diagnoses, which explain the same faulty system. To discriminate among the diagnoses, an additional observation, also called measurement (medical tests or examinations), is needed and the diagnostic procedure must be repeated. The additional observations must be the most promising ones to spare the diagnostic efforts (resources and time). Optimal selection of the proper observation requires a good knowledge of the system and a great deal of previous diagnostic experiences. So, if we want to minimize the diagnostic costs, we would need an experienced user to determine the optimal additional observation. To overcome this requirement we introduce an intelligent diagnostic agent It plays a role of an experienced user and decides upon the further measurement based on previous diagnostic results and knowledge about the system. The agent is very usable for diagnostic applications, where a human solves diagnostic problems in interaction with a model based diagnostic system, what is a very often the case in medical diagnostics. It's main characteristic is that it promptly changes the diagnostic strategy based on previous diagnostic results to minimize diagnostic costs.
机译：基于模型的诊断方法的主要特征是将丰富的知识用于诊断推理。诊断结果是一组有关组件行为的声明，称为诊断，可以解释意外的系统行为。通常，可能会有几种相互竞争的诊断，它们解释了相同的故障系统。为了区分诊断，需要另外进行观察，也称为测量（医学检查或检查），并且必须重复诊断过程。额外的观察结果必须是最有希望的方法，以节省诊断工作（资源和时间）。最佳选择合适的观察值需要对系统有充分的了解，并需要大量的先前诊断经验。因此，如果我们想最大程度地降低诊断成本，则需要有经验的用户来确定最佳的附加观察结果。为了克服此要求，我们引入了智能诊断代理，该代理扮演了经验丰富的用户的角色，并根据先前的诊断结果和有关系统的知识来决定进一步的测量。该代理对于诊断应用非常有用，在该应用中，人员可以与基于模型的诊断系统交互解决诊断问题，这在医学诊断中非常常见。它的主要特点是可以根据以前的诊断结果迅速更改诊断策略，以最大程度地降低诊断成本。

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