推理方法

According to the system view ofdegradation and decomposition put forward above,the paper proposes a newSplit Reasoning Simulation for qualitative model algorithm called SRSQM,which implements the decomposing,reasoning and solving of qualitative models.In another word,it combines variable cluster and causality graph to decompose acomplex qualitative model into many componentsand constraintsamong those components.Through the qualitative reasoning of the componentsand globally propagating the results,the overall solutions can be achieved.

根据复杂系统的递阶分解的系统观点，本文提出一种新的定性模型分解推理仿真算法，称作SRSQM，实现对定性模型的分解推理求解，即对于一个复杂的定性决策模型，应用变量聚类和因果图相结合的方法，将复杂的定性模型分解成若干个单元及其单元之间的约束，通过对各个单元的定性推理以及将单元推理的结果进行全局传播和过滤，得到整个系统的全局解。

Physics concepts, theorems as well as principles are primary content of physics learning, those knowledge take on the forms by proposition . In chapter 3, the study revealed that physics proposition are mainly obtained by reasoning (inductive reasoning such as tallying method, differentiation method and covariant method, deductive reasoning such as negative expression of hypothetical inference, affirmative expression of hypothetical inference ).Then by the means of experiment, the different condition needed when applying those reasoning was explored.

由于定义性概念、定理和定律等命题知识是物理教学的主要内容，所以在第三章中首先通过研究，表明物理命题主要是运用推理策略获得的(归纳推理策略中的求同策略、差异策略和共变策略，演绎推理策略)；然后通过实验等方法研究使用策略需要的条件；最后依据前面的研究提出适合物理命题学习的任务分析方法。

However, the support of stack-based function call is weak in Hoare logic.

此方法源于Hoare逻辑的程序推理以及耶鲁大学的认证汇编编程，它首先需要定义一个推理需要的目标机器和一组静态推理规则，并且证明推理规则的可靠性。

Based on test-score semantics, the canonical method for fuzzy propositions and the organization of fuzzy knowledge base are discussed, and the inference mechanisms of fuzzy expert systems, i.e the fuzzy logic inference for unqualified propositions, the intersection/product syllogism , the consequent conjunction syllogism and the inference of propositional chain ...

基于测试-评分语义学，重点探讨了模糊命题的规范化方法和模糊知识库的组织，研究了模糊专家系统的推理机制：对未限定化命题的模糊逻辑推理；对量化命题的交/积三段论推理，推论连接三段论推理以及命题链的推理。

In ratiocinative mechanism, neural network adopt right directional reasoning method. And the ratiocinative mechanism of expert system contains two conditions, the validation to the diagnosing result of neural network and the alone ratiocination.

在推理机制中，神经网络采用正向推理的方法，专家系统的推理机制则又可分为对神经网络诊断结果的验证和独自重新推理两种情况。

In degraded PSM matching classification, errors modification method of degraded PSM is presented and the matching tensor of every canonical scattering centers is given. In Chapter 4, the theory of wideband millimeter-wave radar target identification is introduced. It is indicated that, for HR radar target identification, description of temoral relations among features and pattern recogntion adaptive to variation of target angles are of key importance. A rule-based pattern recogniton method of sequential reasoning is proposed, which uses a series of rules to describe relations of features variation caused by target angles variation and has the advantages of adaption to unrandom variation and false probability control in classification over traditional statistical pattern recognitiop method. Feature extraction is crucial step in target identification. In range profiles identification application, features are extracted by means of range domain pre-processing algorithm, spacial and amplitude visual computation directly from range profiles and transform algorithm based on range profiles. Visual or transformed features are either sufficiently convinced or necessarily convinced and both of them are effective and robust to range profiles identification.

在第四章，首先对宽带毫米波体制背景下目标识别方法的一些特点进行了阐述并指出，在毫米波雷达目标识别中，特征之间动态关系的描述以及能适应目标姿态角变化的模式识别方法的研究乃是要解决的关键问题；进而提出了一种基于规则库的序贯推理模式识别新方法，在这种方法的研究中，主要包括序贯推理规则库的规则排列与特征选用顺序以及规则库的收敛等问题；这种方法克服了传统的统计模式识别方法特征利用效率不高、难以适应特征值的非随机性变化的缺点，把姿态角变化所引起的特征的变化用一系列规则加以表示，其优点是能适应特征值的非随机性动态变化，并能控制分类过程中的差错概率α；特征抽取是目标识别中的关键步骤，在基于目标距离像的特征抽取方法研究中，提出了距离空间域的预处理算法、距离空间域与幅度域的直观特征抽取方法以及基于目标距离像变换分析的特征抽取方法。

this paper presented a method for improving the inference efficiency,this is achieved by regrouping the knowledge base.

本论文主要提出了一种提高推理效率的方法——知识库重组，这种方法通过缩小推理范围来提高推理效率。

To avoid the collision accidents of ships at sea, with the universalizing and application of computer, and the development and maturity of the technology of Expert System and Artificial Intelligent, this paper developed an intelligent collision avoidance expert system for navigation. The main research work centers on several aspects, which will be represented as below:By the navigation rules" understanding and analyzing, and by the navigation experience and navigation samples" collecting and trimming, we put forth and build a multi-unit and layering KB systematic structure, and implement the KBM. According to features of different knowledge, we adopt multifarious KR, such as: frame KR, production rule KR, procedure KR. We also build a multi-inference system, which based on analog inference, forward illation inference, conversion inference and meta-rule inference. At the same time, we develop each reasoning algorithm. For some problems in collision avoidance region during the building of the expert system, we put forth and build a set of models to solve them using neural network technology.

为了避免船舶间碰撞事故的发生，结合计算机技术在各类船舶中的普及与应用、专家系统技术及人工智能技术的发展、成熟，本文研制、开发了一种具有一定智能的航海避碰专家系统，主要内容包括以下几个方面：通过对航海规则的理解与分析，对航海经验、航海实例的搜集、整理，提出并建立起了航海避碰专家系统的多元分层知识库体系结构，并实现了知识库的管理；根据不同知识的特点，分别采用了框架、产生式规则、过程等多种知识表示方法；提出了一种基于类比推理、正向演绎推理、换位推理及元级推理等的多种推理机制，并建立起了相应的推理算法；引入神经网络技术，针对在建立专家系统过程中所遇到的有关避碰领域内的一些难题，提出并建立了相应解决问题的模型，其中包括：船舶类型的识别，会遇态势的分类及避碰危险的评估。

1Analyzing the various characteristics of quantified structures in Chinese by means of Generalized Quantifier Theory;(2)Extending the function of fragment of natural language. That is, this fragment can not only generate syntactically arguments, including square of opposition and syllogisms, in terms of quantified sentences in Chinese, but interprets semantically the logical validity of those arguments;(3) It is a new study from cognition to generate and interpret the arguments in fragment of natural language.

刻画量化结构及其推理的汉语部分语句系统，其价值如下：（1）运用广义量词理论的方法分析汉语量化结构的各种特征；（2）扩大自然语言部分语句系统的功能，不仅从句法上可以生成由汉语量化句表述的包括对当关系和三段论在内的推理，还能够从语义上解释这些推理句的逻辑有效性；（3）在自然语言部分语句系统内生成并解释推理，这是从认知角度研究推理的新视角。

That is , this fragment can not only generate syntactically arguments , including square of opposition and syllogisms , in ter ms of quantified sentences in Chinese , but interprets semantically the logical validity of those arguments 3 It is a new study from cognition to generate and interpret the arguments in fragment of natural language .

刻画量化结构及其推理的汉语部分语句系统，其价值如下：(1)运用广义量词理论的方法分析汉语量化结构的各种特征；(2)扩大自然语言部分语句系统的功能，不仅从句法上可以生成由汉语量化句表述的包括对当关系和三段论在内的推理，还能够从语义上解释这些推理句的逻辑有效性；(3)在自然语言部分语句系统内生成并解释推理，这是从认知角度研究推理的新视角。

I didn't watch TV last night, because it .

昨晚我没有看电视，因为电视机坏了。

Since this year, in a lot of villages of Beijing, TV of elevator liquid crystal was removed.

今年以来，在北京的很多小区里，电梯液晶电视被撤了下来。