CASIA OpenIR  > 毕业生  > 硕士学位论文
Thesis Advisor戴汝为
Degree Grantor中国科学院自动化研究所
Place of Conferral中国科学院自动化研究所
Degree Discipline模式识别与智能系统
Keyword专家系统 不确定性推理摸型 矛盾的解决 推理结构 不循环覆盖标记逻辑程序 Herbrand支持模型 Expert System Inexact Reasoning Model Conflict Resolution In Ference Structure Acyclic Covered Annotated Logic Program Supp
Abstract分布式人工智能(DAI)系统是一类复杂系统.在这类系统中,由于 知识源的不一致,环境的不确定性和表示的不完全性,矛盾的解决已成 为不可避免的问题.而解决分布式专家系统(DES)在问题求解过程中 产生的结果矛盾则更是DAI领域内一直受到重视的问题.概括地说, 解决这类问题的方法有四种:逻辑方法,逐级上诉推理,协商方法和数 学综合方法.在这四种方法中,虽然数学综合方法能够解决基于不确定 性推理模型的专家系统之间的结果矛盾,但其结论有时与人类的直觉 不符.标记逻辑的最新进展使我们有可能提出新的解决这类矛盾的方 法.支持模型能够为标记逻辑程序给出符合人类直觉的解释,而任何不 循环覆盖程序存在唯一的支持模型.在此基础上我们提出新的解决矛 盾的模型和算法.该方法首先把矛盾的专家系统的推理结构通过一定 的方式结合起来,再引入信息维度对规则重新标记从而构成不循环覆 盖程序,最后通过求支持模型来产生问题的一致解.在很多情况下,这 种方法比数学综合方法更能反映人类的直觉.文章中的例子和结果证 实了该方法的可行性.
Other AbstractDistributed Artificial Intelligence(DAI) system is a kind of complex systems. In these systems, because of the inconsistency of knowledge sources, the uncertainty of environment and the incompleteness of representation, conflict resolution has become an inevitable problem which can be solved with four approaches: logical method, due- process reasoning, negotiation and mathematical synthesis method. Although mathematical synthesis method can deal with the result conflict between expert systems using inexact reasoning model, the final result sometimes is not consistent with our intuition. Recent progress in annotated logic provides a rich setting that allows to disambiguate programs whose rules do not provide enough information to decide which conclusion should be drawn. In this thesis, a conflict resolution model and algorithm between expert systems using inexact reasoning models which aims to solve conflict by combining two inference chains under the setting of annotated logic is proposed. Examples and results are also given.
Other Identifier328
Document Type学位论文
Recommended Citation
GB/T 7714
姜莉楠. 分布式专家系统中解决矛盾的方法[D]. 中国科学院自动化研究所. 中国科学院自动化研究所,1994.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[姜莉楠]'s Articles
Baidu academic
Similar articles in Baidu academic
[姜莉楠]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[姜莉楠]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.