关键词： Internet   搜索工具   决策树算法   粗糙集   最小决策表   网页   链接   机器学习  

摘要： 该论文是安徽省自然科学基金项目\"挖掘internet深层信息的机器学习方法研究\"中的一部分,主要完成的是网页我链接的特征抽取、选优和基于粗糙集的决策树算法的实现以及对特定例子的搜索工具的实现.

关键词： constructive algorithms   divide-and-conquer   neural networks  

摘要： This article presents a machine learning method for solving classification and approximation problems. This method uses the divide-and-conquer algorithm design technique (taken from machine learning models based on a tree), with the aim of achieving design ease and good results on the training examples and allows semi-global actions on its computational elements (a feature taken from neural networks), with the aim of attaining good generalization and good behavior in the presence of noise in training examples. Finally, some results obtained after solving several problems with a particular implementation of SEPARATE are presented together with their analysis.

关键词： 自动文摘系统   文摘句式   机器学习   Internet网  

摘要： 自动文摘系统中一个关键的问题是找出能构成摘要的重点句子。找出这些句子的方法很多，但用机器学习的方法却较少，该文提出了一种关于文摘句式的自动学习方法。该方法以经过简单的预处理的若干语句为训练样本集，以正例句为基点进行由底向上的泛化学习，抽象出关于句式的一般概念，形成句式规则集，作为判断文中哪些语句可作为文摘句的有效手段。这是文摘系统实现的核心部分。

关键词： 机器学习   知识获取   故障诊断   专家系统  

摘要： 首先介绍了故障诊断专家系统中的三种类型的知识 ,即经验知识、因果知识和第一定律知识 ;然后分析了常用的机器学习方法在故障诊断系统的作用 ,并在此基础上将诊断系统中的学习方法分为三种类型 ,即简单学习、交互学习和独立学习 ;最后阐述了故障诊断与机器学习的运行流程及其相互关系 .

关键词： 机器学习   条件概率   层次分类   域   人工智能  

摘要： 本文通过对著名站点Yahoo层次结构的分析，介绍一种基于此层次结构的文档分类器的机器学习技巧。我们利用此层次学习算法来自动构造分类器每个域的背景知识。

关键词： 人工智能   机器学习   智能体   协调   市场   竞争  

摘要： 该文目的是构造一种通用机器学习系统,这种系统集成了多种机器学习方法,适合不同的学习任务,能够适应复杂、大型环境的学习.具体做法是:在系统中设立市场调节机构,资金和知识通过市场进行交换;各学习机采用函数形式表示其活动规律,行动具有主动性并遵循一定的行为规则,目标是资金积累最大化.各学习机购买系统资源进行学习,按照学习得到的知识质量获得报酬,然后进行下一阶段的学习,如此不断循环.在循环过程中,适合的学习机资金积累逐渐增加,活动能力增强,不适合的学习机资金积累逐渐减少,活动性减弱,这样通过优胜劣汰只保留适合的学习方法,系统资源得到了合理分配,并且系统可以达到通用、自适应、自优化的要求.文章最后分析了系统的稳定性.

关键词： 机器翻译   机器学习   搭配规则  

摘要： 本文针对汉英机译系统(CETRAN)中搭配词典的构造,讨论了词语搭配的特点、形式描述以及搭配规则的自动获取问题,提出了一种由词语搭配关系获取机器翻译系统中词类搭配规则的机器学习方法(CRLM)。该方法在对大规模语料进行统计的基础上,找出词语间词类的关联例并引入机器学习技术,构造词类搭配树,获取用于词类有序消歧的搭配规则。

关键词： Crystallization  

摘要： A revised machine learning methodology is applied to a simulated crystallization process flowsheet for continual improvement of its performance by generating and analyzing process data. The aim is to identify bands of crucial decision variables leading to zones of best average process performance. The methodology comprises two components: symbolic induction and case-based reasoning. It uses an incremental algorithm to update performance classification rules, which not only improves the efficiency of the symbolic induction of the classification rules by eliminating the need for their periodic reinduction but also simplifies the case-based reasoning step. The new concepts and procedures are illustrated by application to a potassium nitrate crystallization process comprising a mixed-suspension, mixed-product-removal crystallizer, a hydrocyclone, and a fines dissolver. By identifying and establishing ranges for the most crucial decision variables, namely feed concentration, flow rate, and cooling stream temperature, three zones leaning to an improvement of nearly 12% on nominal average performance are detected within two generations of the classification rules.

关键词： 分类器系统   机器学习   环境标记   遗传算法  

摘要： 该文首先介绍了分类器系统的基本原理,然后对它的一个具体实现-零级分类器系统(ZCS)及其试验环境进行了细致的描述;在此基础上,该文主要完成了三项研究内容:1)针对ZCS结构上的不足进行了相应的改进,使得它在Woods环境下的学习达到了最优值;2)在自然界中存在的\"主动学习\"现象的启发下,抽象出\"环境标记\"的概念,并提出\"基于环境标记的分类器系统\"框架,进而用此框架模型对ZCS进行了扩展;扩展后的ZCS在\"动态Woods\"环境下与其它系统的比较试验结果表明,引入环境标记处理能力对提高分类器系统的学习能力是切实有效的;3)尝试用基于环境标记的分类器系统实现了一个简单的多智能机协作系统.

关键词： DECISION support systems   ARTIFICIAL intelligence   MACHINE theory   MANAGEMENT information systems   MACHINE learning   TELEMATICS  

摘要： When a decision maker has access to multiple expert systems, each embodying a different expert perspective on analyzing and reasoning about the same bind of decision problem, an important consideration is which to use at what times. We address this issue with a method based on competition among the distinct expert systems (and their respective rules). We begin by reviewing prior research concerned with the coordination of multiple sources of expertise in support of decision making, pointing out potential weaknesses of the proposed methods. Next, we introduce a new coordination method based on the competitive paradigm that has been applied in machine learning. This method involves adjustments to the strengths of expert systems and to their constituent rules based on their performances. A nine-step process for adjusting strengths is described. Advantages and limitations of this new method for expert system coordination are discussed. We outline an approach to testing the coordination method and report on preliminary testing of the performance of a system employing our method versus the performance of individual experts.