Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd Logo
Upload

Welcome to Scribd!

  • 1 views

    Ppt1 Introduction Courses

    Uploaded by

    Aditya

    Copyright:

    © All Rights Reserved
    Download as PPT, PDF, TXT or read online from Scribd

    Ppt1 Introduction Courses

    Uploaded by

    Aditya
    1 views8 pages

    Copyright

    © © All Rights Reserved

    Available Formats

    PPT, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Copyright:

    © All Rights Reserved
    Download as PPT, PDF, TXT or read online from Scribd
    Download as ppt, pdf, or txt
    1 views8 pages

    Ppt1 Introduction Courses

    Uploaded by

    Aditya

    Copyright:

    © All Rights Reserved
    Download as PPT, PDF, TXT or read online from Scribd
    Download as ppt, pdf, or txt
    of 8

    ME60353 : Knowledge-Based Systems in Engineering

    (4-0-0) for M.Tech. and Ph.D. students


    MF 41601: Soft Computing
    (3-0-0) for 4th year B.Tech. students

    Dr. D.K. Pratihar


    Professor
    Department of Mechanical Engineering
    IIT Kharagpur
    Email:softcomputinglab@gmail.com
    Website: http://sites.google.com/site/softcomputinglaboratory/Home
    Books
    1.Soft Computing: Fundamentals and Applications By D. K. Pratihar
    New (3-rd) Edition: Soft Computing and AI: Fundamentals and
    Applications
    2.Genetic Algorithms By David Goldberg
    3.Neural Networks By S. Haykins
    4.Fuzzy Sets and Logic By George Klir and others
    Introduction to Knowledge–Based Systems
    • We, human beings, have a natural quest/thirst to know input-output
    relationships of a process/system.

    I1 O1
    I2
    O2
    . Process/system .
    . .
    Im
    On

    Forward Mapping
    O = f(I)
    => [O] = [T][I]
    Reverse Mapping
    [T]-1 [O] = [I]
    Methods
    1. Physics → Mathematics → Differential Equations → Solutions

    2. Real experiments following any statistical DOE → Regression


    analysis

    3. Complex real-world problems → difficult to model mathematically →


    Knowledge-Based Systems/Expert Systems
    What is a Knowledge-Based/an Expert System?
    A Knowledge-based system/an expert system is a computer
    program used to simulate human reasoning for solving a problem

    Notes:
    • An expert system simulates human reasoning about a problem
    domain, whereas ordinary computer program simulates the domain
    itself.
    • Solves the problems using heuristic/approximate methods
    • Does not use any algorithmic and statistical method
    • Consists of a Knowledge Base (KB), an inference engine and
    some forms of user interface
    Inference KB =
    Inputs . . Outputs
    . Engine (DB+RB) .

    •DB: A set of numerical values used to represent the physical


    parameters, which are presented in a particular fashion.

    •RB: Consists of a number of rules, each of which is used to relate


    an output to the inputs using the information of DB.

    •Inference Engine: For a set of inputs, it determines which part of


    the KB is getting fired (becoming active) to determine the output.
    Why do we need an Expert System?
    • To solve complex real-world problems
    • If properly developed, it may give some information, which are
    difficult to foresee beforehand

    Main Task
    To design and develop suitable KB to solve a
    problem
    Syllabus of the course
    Comes under the umbrella of Knowledge Engineering/Applied AI
    We will have to take the help of Soft Computing
    Some Examples of Knowledge–Based Systems

    •DENDRAL: To determine the structure of chemical compounds


    for the given set of constituent elements.
    It could discover a number of unknown structures.

    •MYCIN: To diagnose infectious blood diseases and determine a


    recommended list of therapies for the patient.

    •IBM’s Deeper Blue could defeat the World Chess Champion,


    Gary Kasparov (1997).

    You might also like