Scribd Logo
Upload

Welcome to Scribd!

  • 193 views

    Overview of Expert System Tools

    Uploaded by

    Ranbir Singh
    Expert system tools are designed to support rapid prototyping of expert systems by developing a proof-of-concept program early in the project. Expert system development uses a mix of rapid prototyping and incremental development models rather than a waterfall model. Expert system tools aim to solve integration issues when adding new functionality by using modular knowledge representations. The majority of expert system tools fall into four categories: expert system shells, high-level programming languages, multiple-paradigm programming environments, and additional modules.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PPT, PDF, TXT or read online from Scribd

    Overview of Expert System Tools

    Uploaded by

    Ranbir Singh
    193 views8 pages
    Expert system tools are designed to support rapid prototyping of expert systems by developing a proof-of-concept program early in the project. Expert system development uses a mix of rapid prototyping and incremental development models rather than a waterfall model. Expert system tools aim to solve integration issues when adding new functionality by using modular knowledge representations. The majority of expert system tools fall into four categories: expert system shells, high-level programming languages, multiple-paradigm programming environments, and additional modules.

    Original Description:

    mk

    Original Title

    Overview Of Expert System Tools.ppt

    Copyright

    © Attribution Non-Commercial (BY-NC)

    Available Formats

    PPT, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Expert system tools are designed to support rapid prototyping of expert systems by developing a proof-of-concept program early in the project. Expert system development uses a mix of rapid prototyping and incremental development models rather than a waterfall model. Expert system tools aim to solve integration issues when adding new functionality by using modular knowledge representations. The majority of expert system tools fall into four categories: expert system shells, high-level programming languages, multiple-paradigm programming environments, and additional modules.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PPT, PDF, TXT or read online from Scribd
    Download as ppt, pdf, or txt
    193 views8 pages

    Overview of Expert System Tools

    Uploaded by

    Ranbir Singh
    Expert system tools are designed to support rapid prototyping of expert systems by developing a proof-of-concept program early in the project. Expert system development uses a mix of rapid prototyping and incremental development models rather than a waterfall model. Expert system tools aim to solve integration issues when adding new functionality by using modular knowledge representations. The majority of expert system tools fall into four categories: expert system shells, high-level programming languages, multiple-paradigm programming environments, and additional modules.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PPT, PDF, TXT or read online from Scribd
    Download as ppt, pdf, or txt
    You are on page 1of 8

    Overview Of Expert System Tools

    Expert System Tools : are all designed to support prototyping . Prototype : is a working model that is functionally equivalent to a subset of the product. The idea is to develop, early in the project, a proof of concept program which can be critiqued by user or expert which solves some non-trivial part of the problem. Expert System Development : is often a mix of the rapid prototyping & incremental models of software engineering , rather than water fall model.

    Overview Of Expert System Tools (cont)


    One drawback of the incremental model in more conventional programming paradigms is the problem of integrating new functionality with earlier version . Expert System development environments aim to solve this problem by using modular representations of knowledge (ch 5-8). The majority of software tools for building expert systems seems to fall into 4 broad categories : Expert System Shells . High-level programming languages . Multiple-paradigm programming environments . Additional modules .

    Expert System Shells


    Shells are intended to allow non-programmers to take advantage of the efforts of programmers who have solved a problem similar to their own , thus EMICIN tool allowed the MYCIN architecture to be applied to another medical domain .

    Matching Shells To Tasks


    All shells are not suited to all tasks . Van Melle was among the first to point out that EMYCIN was not a general-purpose problem solving architecture rather he suggested that EMYCIN was suitable for deductive approaches to diagnostic problems where large amounts of data are available and it is possible to enumerate the solution space of diagnostic categories in advance . It is difficult to be rigorous in one's recommendations concerning what shell should be used for what problem .this is because we do not have very clear ideas concerning how the broad range of expert system tasks should be classified .

    Matching Shells To Tasks (cont)


    We shall have more to say about the general problem of selecting an expert system tool in section 17.4 below .With respect to shells , the majority of commercial products initially provided the user with facilities that are only adequate for small search spaces for example exhaustive backward chaining with limited control facilities . Modern shells such as M.4 are claimed to be applicable to a wider range of features for representation & control , such as simulation of forward chaining , procedures , message passing , & so forth .

    Shells & Inflexibility


    The advantegeous simplicity of the knowledge representation language associated with most shells also has a number of disadvantages : The production rule formalism employed by EMYCIN made it difficult to distinguish different type of knowledge , for example : heuristic knowledge , control knowledge , knowledge about expected values for parameters . The relatively unstructured rule set employed by EMYCIN made the acquisition of new knowledge difficult , since adding a rule to the set involved making changes elsewhere in the system , for example : to knowledge tables containing information about medical parameters (this is was the one of the problems that the TEIRESIAS system.)

    Shells & Inflexibility (cont)


    The exhaustive backward chaining employed by EMYCIN as its major mode of inference , involving both meta- & object level rules . Other criticism that is not only the particular implementation of puff in MYCIN . A final criticism of shells concerns the handling of uncertainty . A shell like M.4 comes complete with a particular formalism such as certainty factors for performing inexact reasoning . Thus M.4 runs on PCs under a widely used operating system with database integration &hooks to c, visual BASIC , &visual C++ .

    You might also like