启发式知识获取方法研究

归纳学习是解决知识自动获取的有效方法，针对ID3算法、基于粗集的归纳学习以及其它一些归纳学习方法存在的问题，提出了一种新的归纳学习算法ITIL。此算法用信息增益为启发式，选择尽量少的重要属性或组合，以可分辨性为依据提取规则，许多实例表明，这些规则不仅简单，而且冗余小，作为知识获取模块的一部分，ITIL已被集成到一个“基于知识发现的医疗诊断辅助系统”动态知识库子系统中。     

基于领域语义知识库的疾病辅助诊断方法

健康医疗领域是一个知识密集型的领域,临床诊断的质量主要依赖于医生所掌握的健康医疗知识以及临床经验.然而,单个医生的能力仍然非常有限,所以目前临床诊断的质量并不高.为此,提出一种基于领域语义知识库的疾病辅助诊断方法,基于Freebase中medicine主题域的知识建立了领域语义知识库,提出计算知识库中症状于疾病诊断的权重、计算与患者输入症状集相关的疾病的相关度和基于患者输入症状集推荐相关症状的算法.最后,基于随机选取的6种常见疾病的临床病历数据对所提出的方法与现有方法进行了对比评价,评价结果一方面表明了所提方法对已有方法存在的问题和不足的改进效果,另一方面也表明所提方法可以避免“冷启动”问题,可以快速支撑对大量常见疾病的辅助诊断.基于所提方法,有望为基层全科医生提供大量常见疾病的辅助诊断服务,或者为患者提供疾病自诊服务.     

粗糙集理论在肝病辅助诊断中应用

粗糙集理论主要研究由论域和属性集构成的知识表达系统。医疗诊断中,大量病例、疾病症状和疾病诊断结果构成了一个医学信息决策系统。通过决策属性对条件属性依赖度和重要性分析,发现诊断结果与临床症状之间的关系,提取医学决策规则。实验表明,粗糙集用于肝病辅助诊断方法是正确可行的。     

粗糙集理论在肝病辅助诊断中应用

粗糙集理论主要研究由论域和属性集构成的知识表达系统。医疗诊断中,大量病例、疾病症状和疾病诊断结果构成了一个医学信息决策系统。通过决策属性对条件属性依赖度和重要性分析,发现诊断结果与临床症状之间的关系,提取医学决策规则。实验表明,粗糙集用于肝病辅助诊断方法是正确可行的。     

动态知识智能发现与属性逻辑动态关系

单向S-粗集(one direction singular rough sets)与单向S-粗集对偶(dual of one direction singular rough sets)是S-粗集(singular rough sets)的两种动态结构;在一定条件下,单向S-粗集与单向S-粗集对偶被还原成Z.Pawlak粗集.单向S-粗集与单向粗集对偶分别是S-粗集的基本形式之一.利用单向S-粗集与单向S-粗集对偶,给出动态知识的属性合取范式与属性合取范式萎缩-扩张特征,给出知识推理结构与推理模型.利用单向S-粗集,单向S-粗集对偶,属性合取范式与知识推理交叉、融合、渗透,给出具有属性合取范式萎缩-扩张特征的动态知识生成与生成定理;给出在知识推理条件下的动态知识智能发现与它的属性逻辑关系;给出动态知识的智能筛选、筛选准则、筛选定理与应用.     

基于粗集的工程机械现场故障信息预处理

针对工程机械远程(协同)诊断中现存的由于大量无序数据导致的故障信息传输阻塞、延迟等底层瓶颈问题,从工况监测、故障预测及诊断一体化角度,提出了基于粗集的现场故障信息预处理思想及框架;并在相关粗集知识的基础上,分别提出了装备工况监测和故障诊断的信息约简算法,结合混凝土运输车制动系统给予验证其有效性;在结论中对粗集进一步在故障诊断中的应用进行了探讨.     

一种基于本体的辅助诊断治疗方法

以医学领域本体为基础,为了在医疗领域获取知识优势,结合机器学习技术,采用基于信息熵的属性加权方法结合基于实例的方法对病例进行分类,提出了一种基于本体辅助诊断治疗的方法。对相关工作进行了分析比较,给出了一个辅助诊断治疗心肌梗塞的例子,以说明该方法的有效性。与一般的医疗诊断方法相比,具备更良好的共享能力和扩展能力,可提高辅助医疗诊断、治疗的准确性。最后给出了结论和下一步工作。     

基于广义粗集覆盖约简的粗糙熵

在广义粗集覆盖约简理论中,由于集合的上下近似是由其覆盖约简来确定的,因此有必要寻求一种新的度量来刻画知识和粗集的粗糙性.通过引入信息熵以刻画广义粗集覆盖约简的知识粗糙性以及粗集粗糙性,提出了一种新的知识粗糙性和粗集粗糙性度量.得到知识粗糙熵和粗糙集的粗糙熵都随广义覆盖约简的变细而单调减少的结论,从信息论观点出发,对不完备信息系统粗集理论进行了探讨.     

软件密集型装备软件故障诊断技术研究

针对当前软件密集型装备软件故障诊断方法的缺陷,提出了一种基于粗诊断与细诊断技术相结合的软件故障定位方法。对故障进行初步定位,找出软件故障可疑模块集,利用切片技术对故障模块集进行检测,将故障定位到程序语句集中,采用代码分析辅助诊断工具对可疑代码集进行排查、定位。软件密集型装备故障诊断原型系统的应用表明,该方法能有效地进行软件故障的定位。     

基于特征融合的产科多标记辅助诊断研究

中文产科电子病历中蕴含着大量的医疗知识和健康信息,电子病历的信息抽取及辅助诊断对提高人口的生育健康水平具有重要意义。电子病历中,首次病程记录的入院诊断是根据主诉、辅助检查、查体等信息得出的。通常情况下诊断中包含正常诊断、病理诊断及并发症而非单一结果。因此,该文将辅助诊断问题转化为多标记分类任务。在对产科电子病历的首次病程记录进行数据清洗和结构化后,规范化诊断结论,将LDA所抽取的文本特征与病历中的数字特征采用向量拼接的方法融合为新的特征,再按诊断结果出现的频次不同形成不同的多标记集,根据首次病程中部分信息进行辅助诊断,采用RAkEL、MLkNN、CC和BP-MLL方法进行多标记分类。实验结果表明,采用融合特征的多标记分类方法,能够提升中文产科电子病历辅助诊断的效果。     

A Software Tool for Constructing Traditional Chinese Medical Expert Systems

It is an urgent task to hnplemeut a lot of expert systems to capture the valuable expertise ofexperienced doctors of traditional Chinese medicine.In order to meet the needs,a software tool isdeveloped.It features a unified diagnosis model,a specially designed knowledge representationlanguage and an efficient but effective inference engine.To implement an expert system,it isonly necessary to input the expert's knowledge expressed in knowledge representation languagewithout the design of any additional software.The time and effort required for implementing anexpert system are thus greatly saved.The software is very compact and can run onmicrocomputers e.g.IBM-PC/XT.Two traditional Chinese medical expert systems have beensuccessfully implemented with the tool.     

医疗诊断专家系统推理机的设计与实现

专家系统是人工智能领域的重要分支,推理机是专家系统的重要组成部分。文中用关系数据库SQLServer2000设计诊断知识库,构造了一个医疗诊断专家系统,其推理机能够有效模拟医生的诊断思维。因此,它可以作为医生诊断疾病的一种辅助工具。根据疾病诊断的要求,仅对医疗诊断推理机的设计与实现方法进行了探讨,提出了一种基于正-反向推理的精确与不精确推理相结合的推理策略,并采用关系数据库和面向对象的技术来具体编程实现,从而提高了推理机的效率,获得了较好的推理效果。     

An expert system for federal tax planning

Abstract: Recent developments in a subarea of computer science called artificial intelligence have included the creation of expert systems that are capable of solving difficult applications problems which require expert knowledge for their solution. Such expert systems have been found to be useful in a number of applications (e.g. medicine, biochemistry and mineral exploration). In this paper the author presents an expert system for solving problems concerning income and transfer tax planning for individuals In developing this system, a theoretical structure and a set of decision rules were specified and then programmed into a rule-based system that had previously been used for medical diagnosis (Mycin [1]) Once the system was developed, its problem-solving capabilities were refined and verified by a panel of tax experts using a blind verification procedure. This verification step demonstrated that an expert system could be developed in that domain.     

A fuzzy expert system approach for real-time monitoring of endemic diseases

In this paper, an architecture of LEPDIAG—a knowledge based system for on-line diagnosis and for monitoring prognosis of leprosy—is presented. The important features of LEPDIAG that have been detailed are a multiple expert environment, a homeostatic expert containing the model of immune reaction, a performance evaluator that can compare the observed signs and symptoms with those predicted by homeostatic expert, and a prognostic expert which optimizes the management schedule for the patients. The entire system is built around a fuzzy expert system building tool FRUIT to deal with the imprecise knowledge.     

基于Multi-Agents 分布式医学诊断系统研究

医学诊断系统是一个新兴的复杂的应用系统,人工智能技术,计算机协作支持技术及 高速通信网络体系结构的发展促进了计算机支持的诊断系统的发展．当前医学诊断系统的难 点在于如何利用网络这个资源分布平台来获取所需要的数据及在数据不完整状态进行推理求 解，而这些问题的解决在于能够有一种机制使得能在一个标准的应用系统结构中准确的表示 并获取信息及集成各种医学资源使之相互协作．本文描述了一种利用多智能体(Multi-agen ts system, MAS)体系结构和中间件(middleware)技术如公共请求代理结构(Common Object Request Broker Architecture, CORBA)进行设计的分布式医学诊断系统. 该系统能集成多 种医学资源和医学应用实体并且能实现参与诊断的医学实体之间的协作，以减少由于信息缺 乏而带来的诊断偏差. 另外本文还将一种实验室开发的模糊最小最大神经网络(Fuzzy Min- Max Neural Network, FMMNN)的模糊规则提取方法应用于该系统以证实该分布式诊断系统的 优越性.     

基于知识库的企业诊断原型系统的设计与实现

企业诊断是改善企业管理工作,提高企业经济效益的有效手段。利用计算机获取专家诊断知识,自动智能地分析企业数据,实时对企业作出诊断与评价,对企业发展有着深元的影响,文中研究了制造类型企业的诊断方法和技术,建立了企业诊断和知识库的系统结构,并设计与实现了企业诊断计算机软件原型系统。     

Knowledge-based understanding of radiology text

A data acquisition tool which will extract pertinent diagnostic information from radiology reports has been designed and implemented. Pertinent diagnostic information is defined as that clinical data which is used by the HELP medical expert system. The program uses a memory-based semantic parsing technique to 'understand' the text. Moreover, the memory structures and lexicon necessary to perform this action are automatically generated from the diagnostic knowledge base by using a special purpose compiler. The result is a system where data extraction from free text is directed by an expert system whose goal is diagnosis.     

A tool for task-based knowledge and specification acquisition

Knowledge acquisition has been identified as the bottleneck for knowledge engineering. One of the reasons is the lack of an integrated methodology that is able to provide tools and guidelines for the elicitation of knowledge as well as the verification and validation of the system developed. Even though methods that address this issue have been proposed, they only loosely relate knowledge acquisition to the remaining part of the software development life cycle. to alleviate this problem, we have developed a framework in which knowledge acquisition is integrated with system specifications to facilitate the verification, validation, and testing of the prototypes as well as the final implementation. to support the framework, we have developed a knowledge acquisition tool, TAME. It provides an integrated environment to acquire and generate specifications about the functionality and behavior of the target system, and the representation of the domain knowledge and domain heuristics. the tool and the framework, together, can thus enhance the verification, validation, and the maintenance of expert systems through their life cycles. © 1994 John Wiley & Sons, Inc.     

Knowledge-based expert systems in offshore structural design

With the world growing more complex, computers must help not only by doing ordinary computing but also by doing computing that exhibits intelligence. Efforts in the application of Artificial Intelligence (AI) to intelligent problem solving led to the development of expert systems. An expert system is a useful tool for solving ill defined problems that require a great deal of specialized knowledge that experts in a particular field acquire from long experiences with such tasks. The expert knowledge is contained in what is aptly called a knowledge base.

Offshore structural design is significantly influenced by expert human knowledge. Much of the knowledge is heuristic and attained through experience over the years. The objective of this paper is to highlight the possible use of expert systems in the area of offshore structures. A brief overview of knowledge systems followed by an application example is presented.