1) fuzzy rule extraction
模糊规则抽取
3) the extraction of fuzzy rules
模糊规则获取
1.
Since the final membership function value,if obtained by extracting fuzzy rules on high dimension space(face,for instance),is not so reasonable and might very well be zero values,human adjustment is made on the initial value of the membership function during the extraction of fuzzy rules.
对基于模糊神经网络的人脸识别方法进行了研究,提出的模糊神经网络在高维空间(如人脸)上进行模糊规则获取中得到的最终隶属度函数值并不合理,易得到零值的情况,对模糊规则获取过程中的隶属度函数初始值进行人为地调整。
4) rule extraction
规则抽取
1.
Text categorization rule extraction based on fuzzy decision tree;
基于模糊决策树的文本分类规则抽取
2.
A web document categorization rule extraction based on chaos particle swarm optimization combining linkage clustering;
结合链接结构聚类的混沌粒子群网页分类规则抽取
3.
Study on Attributions Selection and Rule Extraction of Data Mining for Classification Based on Neural Networks;
基于神经网络的分类数据挖掘属性选择和规则抽取研究
5) Extraction rules
抽取规则
1.
This method generates information extraction rules that is based on DOM path through attaching syntax information and sample-learning.
该方法通过附加语义、样本学习生成基于DOM路径的抽取规则,利用遍历DOM树实现信息抽取。
2.
In the framework, the autonomy and coordination of agent are employed to assist user formulate extraction queries, to learn extraction rules on the base of knowledge bases and so on.
在这个原型系统里,利用Agent的自治能力和合作能力来协助用户对抽取请求进行公式化表述和结合知识库学习抽取规则等。
3.
The program that extracts information from web is called wrapper, and the main task of constructing wrapper is to prepare extraction rules.
从网页中抽取信息的程序叫包装器(Wrapper),构建包装器的主要任务是编写抽取规则,因此,编写健壮灵活的抽取规则成为信息抽取的研究重点。
6) extraction rule
抽取规则
1.
Web Information Extraction Rules and Their Learning Algorithms;
Web信息抽取规则及其学习算法
2.
Visualized implementation of Web information extraction rule based on mini-node DOM tree
最小节点信息树抽取规则及可视化生成方法
3.
In this paper,we mainly discussed DOM-based Web information extraction,studied how to construct extraction rules to improve precision ratio of extraction and adaptation of extraction rules,and the rules generation procedure is also presented.
本文主要讨论基于DOM的Web信息抽取,研究如何构造抽取规则,才能提高信息抽取的准确度、提高抽取规则的适应能力,并给出了抽取规则的生成过程。
补充资料：模糊规则
模糊规则的形式为：if x is A then y is B
其中A和B为由论域X和Y上的模糊集合定义的语言值。“x is A”称为前提，“y is B”称为结论。
以上模糊规则可以简写为A → B。本质上模糊规则是定义在X × Y上的二元模糊关系R。A → B有
两种解释，一种是A耦合（coupled with）B:
<math>R=A \rightarrow B = A \times B = \int_{X \times Y}^{}tnorm(\mu_A (x), \mu_B (y))/(x,y)</math>
另一种是A导致（entails）B：
<math>R=A \rightarrow B = \bar{A} \cup B</math>
基于以上两种解释和不同的tnorm, tconorm算子，模糊规则可以有多种合法的计算公式。
说明：补充资料仅用于学习参考，请勿用于其它任何用途。
参考词条