ExpertSystem 专家系统_第1页
ExpertSystem 专家系统_第2页
ExpertSystem 专家系统_第3页
ExpertSystem 专家系统_第4页
ExpertSystem 专家系统_第5页
已阅读5页,还剩40页未读 继续免费阅读

下载本文档

版权说明:本文档由用户提供并上传,收益归属内容提供方,若内容存在侵权,请进行举报或认领

文档简介

1、Expert systems are computer programs built for commercial application using the programming techniques of AI, especially for problem solving.Expert systems have been built for a variety of purposes including medical diagnosis, electronic fault finding, mineral prospecting, and computer system config

2、uration. Expert systems which closely matches the human logical thinking process.Important features of expert system:Facility for non-expert personnel to solve problems that requires some expertise.Speedy solutions.Reliable solutionsCost reductions.Elimination of uncomfortable and monotonous operati

3、ons.Power to manage without human experts.Wider access to knowledge.Experts systems are recommended when:Human experts are difficult to find.Human experts are expensive.Knowledge improvement is needed.Knowledge is difficult to acquire and is based on rules that can only be learnt through experience.

4、The available information is poor, partial, incomplete.Problems are incomplete defined.There is a lack of knowledge among all those who need it.The problem is subject to rapidly changing legal rules and codes.UserThe user interface may employ the following style:Question and answerMenu drivenNatural

5、 languageGraphic interfaceKnowledge base editorInference engineExplanation subsystemGeneral knowledge baseCase-specific dataExpert System OrganizationExpert systems are generally,Open to inspection, both in presenting intermediate steps and in answering questions about the solution process.Easily mo

6、difiable, by both adding and deleting skills from the knowledge base andHeuristic in using (often imperfect) knowledge to obtain solutions.We can represent the knowledge of an expert system using rules and objects.Knowledge is represented in the form of IF-THEN rules as follows:IF the load demand is

7、 medium THEN the system is reliableRules can pattern-match on objects as well as facts.An expert system is a collection of programs or computer software that solve problems in the domain of interestdomain of interest. It is called a system because it consists of both a problem solving component and

8、a support component.Development of expert system involving human experthuman expert, knowledge engineerknowledge engineer and knowledge base of expert systemknowledge base of expert system.Human ExpertKnowledge EngineerKnowledge base of expert systemDevelopment of an expert systemThe process of buil

9、ding an expert system is called knowledge engineering and is done by a knowledge engineer.Knowledge engineer:Background in computer science and know how to build an expert system.Decides how to represent the knowledge in an expert system.Help the programmers to write the code.Gather/acquire knowledg

10、e from human expert or any other source.The expert will evaluates the expert system and gives report to the knowledge engineer.An explanation facility is an integral part of a sophisticated expert system.A practical limitation of many expert systems is lack of general knowledge. Heuristics are not g

11、uaranteed to succeed in the same way that an algorithm is a guaranteed solution to a probem.The basic characteristics required for an expert system are:High performanceExpertiseAdequate response timeGood reliabilitySelf knowledgeUnderstandableJustificationFlexibilityUserUser InterfaceKnowledge baseI

12、nference EngineFacts, queriesAdvice, consultation and justificationArchitecture of an expert systemThe user interfaceuser interface allows the system user to enter rules and facts about a particular situation and ask questions of the system, provides response to user requests, and supports all other

13、 communication between the system and the user.The knowledge of a human expert on a particular subject is contained in codified form within the KB.The inference engine uses the information provided to it by the KB and the user to infer new facts.This procedure can simulate the deductive thought proc

14、esses of an expert.The expert systems that represent knowledge in a rule format are known as rule-based expert systemrule-based expert system. Goals of expert systems:Substituting for an unavailable human expert, combining the knowledge and experiences of several human experts, training new experts,

15、 providing requisite expertise on projects that do not attract or retain experts, providing expertise to projects that cannot afford experts.Their basic activities can be grouped into different categories:InterpretationPredictionDiagnosisDesignPlanningMonitoringDebuggingRepairInstructionControlCateg

16、oryCategoryProblem addressedProblem addressedInterpretationInferring situation descriptions from sensor dataPredictionInferring the likely consequences of given situationsDiagnosisInferring system malfunctions from observationsDesignConfiguring objects under constraintsPlanningDesigning actionsMonit

17、oringComparing observations to expected outcomesDebuggingPrescribing remedies for malfunctionsRepairExecuting plans to administer prescribed remediesInstructionDiagnosis, debugging and repairingControlGoverning overall system behaviourSome of the advantages of using expert systems are the following:

18、Increased availabilityReduced costReduced dangerPermanenceMultiple expertiseIncreased reliabilityExplanationFast responseSteady, unemotional, and complete response at all timeIntelligent tutorIntelligent databaseCONVENTIONAL PROGRAMEXPERT SYSTEMObjective: to perform some tasks and is expected to per

19、form the tasks correctly or get the correct answers every time.Algorithmic activitiesIt is not expected to give the right answer all the time.No algorithmic solution Rely on inferences to achieve a reasonable solutionIt can explain the criteria used and the conclusions reached on the way to arriving

20、 at the solution.Human expertsHuman expertsExpert systemsExpert systemsConventionalConventional programs programsUse knowledge in the form of rules of thumb or heuristics to solve problems in a narrow domain.Process knowledge expressed in the form of rules and use symbolic reasoning to solve problem

21、s in a narrow domain.Process data and use algorithms, a series of well-defined operations, to solve general numerical problems.In a human brain, knowledge exists in a compiled form.Provide a clear separation of knowledge from its processing.Do not separate knowledge from the control structure to pro

22、cess this knowledge.Capable of explaining a line of reasoning and providing the details.Trace the rules fired during a problem-solving session and explain how a particular conclusion was reached and why specific data was needed.Do not explain how a particular result was obtained and why input data w

23、as needed.Use inexact reasoning and can deal with incomplete, uncertain and fuzzy information.Permit inexact reasoning and can deal with incomplete, uncertain and fuzzy data.Work only on problems where data is complete and exact.Can make mistakes when information is incomplete or fuzzy.Can make mist

24、akes when data is incomplete or fuzzy.Provide no solution at all or a wrong one, when data is incomplete or fuzzy.Enhance the quality of problem solving via years of learning and practical training. This process is slow, inefficient and expensive.Enhance the quality of problem solving by adding new

25、rules or adjusting old ones in the knowledge base. When new knowledge is acquired, changes easy to accomplish.Enhance the quality of problem solving by changing the program code, which affects both the knowledge and its processing, making changes difficult.Source: Michael Negnevitsky, 2005, Artifici

26、al Intelligence: A guide to intelligent systems, Addison Wesley.Some of the main steps to be followed in designing and developing an expert system:Statement of the problem to be solvedSearch for the expert system or the equivalent data or experienceDesign of the expert systemSelection of the degree

27、of participation of the user.Selection of the development tool, shell, or programming language for development.Development of a prototype.Prototype checking.Refinement and generalizationMaintenance Updating.Experts knowledge errorsSemantic errorsSyntax errorsInference engine errorsInference chain er

28、rorsLimits of ignorance errorsExpert errorKnowledge engineer errorKnowledge base errorInference engine errorInference chain errorMajor errors in expert systemUsing rule-based systems is the most popular way to represent knowledge within an expert system.They are less expensive, and many expert syste

29、m development packages employ rule bases.The primary people involved in building an expert system are the knowledge knowledge engineerengineer, the domain expertthe domain expert and the end the end useruser.Expert SystemEnd-userKnowledge EngineerProgrammerDomain ExpertProject ManagerExpert SystemDe

30、velopment TeamThe core of an expert system is the knowledge baseknowledge base.The knowledge base is the set of production rules.In a rule-based system, these condition-action pairs are represented as rules, with the premises of the rules (the IF) corresponding to the condition and the conclusion (t

31、he THEN) corresponding to the action.The inference engine is the recognize-act part of the production system.It carries out the reasoning whereby the expert system reaches a solution.It links the rules given in the knowledge base with the facts provided in the database.It may be either data-driven (

32、forward chaining) or goal-driven (backward chaining).Inference EngineKnowledge BaseRule: IF-THENDatabaseFactExplanation FacilitiesUser InterfaceUserKnowledge BaseDatabaseFact: A is xMatchFireFact: B is yRule: IF A is x THEN is yRule 1:IFY is trueANDD is trueTHEN Z is trueRule 2:IFX is trueANDB is tr

33、ueANDE is trueTHEN Y is trueRule 3:IFA is trueTHEN X is trueAXBEYDZMatchFireKnowledge BaseDatabaseABCDEXMatchFireKnowledge BaseDatabaseABCDELXMatchFireKnowledge BaseDatabaseACDEYLBXMatchFireKnowledge BaseDatabaseACDEZYBLXCycle 1Cycle 2Cycle 3X & B & EYZY & DLCL & MAXNX & B &

34、EYZY & DLCL & MAXNX & B & EYZY & DLCL & MAXNX & B & EYZY & DLCL & MAXNSub-Goal: XSub-Goal: YGoal: ZMatchFireKnowledge BaseDatabaseABCDEXMatchFireKnowledge BaseDatabaseACDEYXBKnowledge BaseDatabaseACDEZYBXMatchFirePass 2Knowledge BaseGoal: ZKnowledge BaseSub-Go

35、al: YKnowledge BaseSub-Goal: XPass 1Pass 3Pass 5Pass 4Pass 6DatabaseABCDEDatabaseABCDEDatabaseBCDEAYZ?X?X & B & EYLCL & MAXNZY & DX & B & EYZY & DLCL & MAXNLCL & MNX & B & EYZY & DAXX & B & EYZY & DLCL & MAXNX & B & EYLCL &

36、MAXNZY & DX & B & EYZY & DLCL & MAXNAn expert is a person who has deep knowledge in the form of facts and rules and strong practical experience in a particular domain. An expert can do things other people cannot.Production rules are represented as IF (antecedent) THEN (consequent) statements.A production rule is the most popular type of knowledge representation.A computer program capable of performing at a human-expert level in a narrow problem domain area is called an expert system. The most popular exper

温馨提示

  • 1. 本站所有资源如无特殊说明,都需要本地电脑安装OFFICE2007和PDF阅读器。图纸软件为CAD,CAXA,PROE,UG,SolidWorks等.压缩文件请下载最新的WinRAR软件解压。
  • 2. 本站的文档不包含任何第三方提供的附件图纸等,如果需要附件,请联系上传者。文件的所有权益归上传用户所有。
  • 3. 本站RAR压缩包中若带图纸,网页内容里面会有图纸预览,若没有图纸预览就没有图纸。
  • 4. 未经权益所有人同意不得将文件中的内容挪作商业或盈利用途。
  • 5. 人人文库网仅提供信息存储空间,仅对用户上传内容的表现方式做保护处理,对用户上传分享的文档内容本身不做任何修改或编辑,并不能对任何下载内容负责。
  • 6. 下载文件中如有侵权或不适当内容,请与我们联系,我们立即纠正。
  • 7. 本站不保证下载资源的准确性、安全性和完整性, 同时也不承担用户因使用这些下载资源对自己和他人造成任何形式的伤害或损失。

评论

0/150

提交评论