归纳规则

Conditions of consistency and closure are necessary conditions that should be satisfied by knowledge set.

根据什么规则接受归纳结论是归纳逻辑研究中的一个重要问题。

But the two bases are treated much differently in traditional textbooks. This article noted several striking features about the design of this English coursebook that merits our attention:(1) Grammar learning process is no longer done dryly in the traditional way.

学生受到激发后，自己归纳出语言规则，而且能在有趣的语言上下文中能充分理解语言规则。2、在爽心悦目的课堂活动中，轻松的语言输入过程在不知不觉中带来了学习者的语言输出。

This paper research and compare board game has simultaneously two people and three people ,like Chinese checkers, Abalone, and Chinese chess.Generalize little difference from its rules and ways, them increases rules in order to fair. This research attempt to revises and publish the feasible idea an example of Hex. First I simplify the checkerboard to increase the universality, and two opposite sides expand for three opposite to reduce first superiority, and make the checkerboard game fair.Besides, increases a player becomes three player of checkerboard games and increases interest. Also develops the new computer checkerboard game of Hex by the C# Windows programming language, and provides a new playing chess the game platform.

本论文研究及比较同时有双人及三人玩法规则的棋盘游戏─跳棋、大王鲍棋、象棋，归纳发现其规则及玩法差异不大，主要为了公平性加入其他规则；本文将以 Hex棋为例，尝试修改，提出可行的办法，简化棋盘以增加普遍性，两对边扩充为三边，以减少先下者的优势，使棋盘游戏更公平，除双人对弈外，同时也增加下棋人数，成为三人棋盘游戏，增加趣味，并且以C#视窗程式语言开发新Hex棋成为电脑棋盘游戏，提供一个新的对弈的游戏平台。

They are hidden in legal rules, while to find the media inside, we have to sum up, hackle, and explain in detail.

这些都蕴含于法律规则中，而要寻求其中的中介或载体，就需要详尽地、分门别类地归纳、梳理和阐释相关规则。

In this framework for rule induction,the authors particularly focus on the experiments with the WHISK algorithm for generating patterns.Experimental results show that the system performs well on both single-slot and multi-slot extraction tasks.

在用于抽取规则归纳的框架下，还重点对用于生成抽取模式的WHISK学习算法进行了实验对比分析，结果表明系统对于单槽和多槽数据都具有不错的归纳学习能力。

ID〓 is characterized by the fast speed of training and classifying, and it can be applied to large-scaled learning problems. But the selection criterion of attribute is the core of ID〓, and is now doubted. AQ〓 is a type of decision rule induction, presented by Michalski, a scientist of U. S.. It is characterized by high-precision classifying and capability of knowledge expression, and it can be applied to automatic knowledge acquisition in expert system. However, the inducing and matching speed of AQ〓 is slow.

其中决策树归纳以澳大利亚学者Quinlan的ID〓为代表，其特点是训练与分类速度都很快，适用于大规模的学习问题，但ID〓把注意力集中在属性的选择标准上，然而这种思想方法已开始受到人们的怀疑；决策规则归纳以美国学者Michalski的AQ〓为代表，其特点是分类精度高和知识表达能力强，适用于专家系统的知识自动获取，但归纳和匹配速度都慢。

For multi-layer fabric, Kronecker product and yarn grouping of different layers according to the arrangement ratio are combined together to explore the mathematical model of weaves automatic generation. The qualifications of forming the multi-layer weaves are also discussed.

对单系统织物组织类别，首先根据数学描绘的需要进行了重新划分，分为规则组织、准规则组织和非规则组织，接着分别建立了各种类别组织的数学模型，并对组织自动生成的参数和组织构成条件进行了归纳和总结；对于多系统组织的重组织和层组织，结合不同系统纱线按排列比的分组法和矩阵的Kronecker叉积运算建立了组织结构矩阵自动生成的数学模型，并对多系统组织构成条件的数学模型进行了分析。

For single-layer fabric, weaves are re-grouped for the purpose of mathematical description, and they fall into three groups: regular weaves, pre-regular weaves and irregular weaves. And then the mathematical models of every group are established, the parameters and the qualifications of weaves automatic generation are offered too. For multi-layer fabric, Kronecker product and yarn grouping of different layers according to the arrangement ratio are combined together to explore the mathematical model of weaves automatic generation. The qualifications of forming the multi-layer weaves are also discussed.

对单系统织物组织类别，首先根据数学描绘的需要进行了重新划分，分为规则组织、准规则组织和非规则组织，接着分别建立了各种类别组织的数学模型，并对组织自动生成的参数和组织构成条件进行了归纳和总结；对于多系统组织的重组织和层组织，结合不同系统纱线按排列比的分组法和矩阵的Kronecker叉积运算建立了组织结构矩阵自动生成的数学模型，并对多系统组织构成条件的数学模型进行了分析。

The process features and shape features of the electronic part are classified and defined, and then the feature model library is built, which enables rapid input of the part information. The rules and methods of stamping process design are summarized to build the stamping process library, so that the stamping process can be designed rapidly and intelligently. Furtherly, the process design rules and methods in process station layout of multi-station progressive die as well as the methods of decompounding and compounding process and the expression method of the process station layout are concluded and summarized, and then the rule library and case library of process station layout are built. Designing the process station layout rapidly is carried out based on rules and cases. At last the rapid design function of process station layout is validated through simulating the rapid designing layout system of multi-station progressive die on Pro/E 3D platform.

主要研究成果有：对电子产品冲压件的工艺特征与形状特征进行了分类和定义，建立了特征模型库，实现了冲压件信息的快速输入；归纳了冲压工艺设计规则和方法，建立了冲压工艺知识库，实现了冲压工艺快速与智能化设计；归纳总结了多工位级进模工步排样中的工序设计原则和方法，工序分解与工序组合方法，工步排样知识的表示方法，建立了工步排样规则库和实例库，利用基于规则和实例的推理方法，实现了工步排样的快速设计；在Pro/E三维软件平台上开发了电子产品多工位级进模工步排样快速设计系统，系统通过实例验证了工步排样的快速设计功能。

Inductive learning is one of the most important,the corest and the maturest branches in machine learning, but for using the knowledge acquired by learning and improving inductive learning algorithms, there are a lot of problems that are hard to solve using traditional methods.

1引言机器学习是当前人工智能的一个热门学科，归纳学习旨在从大量经验数据中归纳抽取一般的判定规则和模式，是机器学习最重要、最核心也是最成熟的一个分支。

The absorption and distribution of chromium were studied in ryeusing nutrient culture technique and pot experiment.

采用不同浓度K2CrO4(0，0.4，0.8和1.2 mmol/L)的Hoagland营养液处理黑麦幼苗，测定铬在黑麦体内的亚细胞分布、铬化学形态及不同部位的积累。

By analyzing theory foundation of mathematical morphology in the digital image processing, researching morphology arithmetic of the binary Image, discussing two basic forms for the least structure element: dilation and erosion.

通过分析数学形态学在图像中的理论基础，研究二值图像的形态分析算法，探讨最小结构元素的两种基本形态：膨胀和腐蚀；分析了数学形态学复杂算法的基本原理，把数学形态学的部分并行处理理念引入到家实际应用中。