首页> 外文会议>International Joint Conference on Artificial Intelligence >An Evidence-based Framework for a Theory of Inheritance
【24h】

An Evidence-based Framework for a Theory of Inheritance

机译:基于证据的继承理论的框架

获取原文
获取外文期刊封面目录资料

摘要

We present an approach to formalizing nonmonotonic multiple inheritance networks by combining concepts from logic programming and multi-valued logics in a uniform framework. A Horn-clause logic language is used for specifying inheritance networks. This allows a natural representation of class-subclass hierarchies and ambiguous inheritance networks. It also provides means for resolving ambiguities resulting from the network topology, but which are not inherent to the problem. We provide a model theory for the language and show how a unique intended model can be associated with every inheritance network. This model resembles the unique extension obtained in the skeptical theory of inheritance [Hor-87], but is more general. Finally, we present an algorithm which realizes the aforementioned semantics.
机译:我们通过在统一框架中将逻辑编程和多值逻辑的概念组合来表示非单调多重继承网络的方法。 Horn-ClareAse逻辑语言用于指定继承网络。这允许类子类层次结构的自然表示和含糊的继承网络。它还提供了解决由网络拓扑产生的含糊不清的含糊不清的方法,但这不是固有问题。我们为语言提供了模型理论,并展示了如何与每个继承网络相关联的独特预期模型。该模型类似于在遗产持怀疑理论中获得的独特扩展[Hor-87],但更为一般。最后,我们提出了一种实现上述语义的算法。

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