expert system and decision making

download expert system and decision making

of 22

Transcript of expert system and decision making

  • 7/29/2019 expert system and decision making

    1/22

    EXPERT SYSTEM

    AND DECISION

    MAKING

    - Parul Jain

  • 7/29/2019 expert system and decision making

    2/22

    INTRODUCTION

    Artificial intelligence-based computerprograms called Expert Systems havereceived a great deal of attention.

    They are used in a huge variety of fields

    like medicine, mathematics, engineering,computer science and education etc.

    An expert system stores the knowledgeof one or more human experts in a

    particular field. ES are used for problems for which there

    is no single "correct" solution which canbe encoded in a conventional algorithm.

  • 7/29/2019 expert system and decision making

    3/22

    EXPERT SYSTEM

    COMPONENTS The part of the expert system that

    stores the knowledge is called the

    knowledge base.

    The part that holds the specifics of theto-be-solved problem is called the

    global database

    The part that applies the knowledge tothe problem is called the inference

    engine.

  • 7/29/2019 expert system and decision making

    4/22

    How to build an ES?

    A knowledge engineer starts byreading domain-related literature tobecome familiar with the issues and

    the terminology. holds extensive interviews with one or

    more domain experts to "acquire" theirknowledge.

    organizes the results of theseinterviews and translates them intosoftware that a computer can use.

  • 7/29/2019 expert system and decision making

    5/22

    How to build an ES? (cont)

  • 7/29/2019 expert system and decision making

    6/22

    CHARACTERSTICS

    Operates as an interactive system Tools have ability to filter knowledge

    Make logical inferences based on

    knowledge storedAbility to Explain Reasoning

    Domain-Specific

    Applications Cost-effective alternative to Human

    Expert

  • 7/29/2019 expert system and decision making

    7/22

    FEATURES

    Goal driven reasoning or backwardchaining

    Coping with uncertainty

    Data driven reasoning or forwardchaining

    Data representation

    User interface

    Explanations

  • 7/29/2019 expert system and decision making

    8/22

    EXPERT SYSTEMS ANDARTIFICIAL INTELLIGENCE

    AI programs that achieve expert-levelcompetence in solving problems in task areasby bringing to bear a body of knowledgeabout specific tasks are called knowledge-basedorexpert systems.

    The area of human intellectual endeavour tobe captured in an expert system is called thetask domain.

    Taskrefers to some goal-oriented, problem-

    solving activity. Domainrefers to the area within which the

    task is being performed.

    The Turing Test

  • 7/29/2019 expert system and decision making

    9/22

    THE WORKING OF ES

    Every expertsystem consists of

    two principal parts:

    the knowledge

    base; and the

    inference engine.

    The knowledge

    baseof expertsystems contains

    both factual and

    heuristic

    knowledge.

  • 7/29/2019 expert system and decision making

    10/22

    The working of ES (cont.)

    Factual knowledgeis that knowledge

    that is widely shared found in

    textbooks or journals, and agreedupon by those knowledgeable in the

    particular field.

    Heuristic knowledgeis the lessrigorous, more experiential, more

    judgmental knowledge of

    performance.

  • 7/29/2019 expert system and decision making

    11/22

    KNOWLEDGEREPRESENTATION

    Knowledge representation formalizes andorganizes the knowledge.

    One representation is a ruleconsisting of an IF partand a THEN part.

    Another representation, called the unitis based

    upon a more passive view of knowledge. The problem-solving modelcontrols the steps

    taken to solve the problem. forward chaining -If the chaining starts from a set of

    conditions and moves toward some conclusion, the

    method. backward chaining -If the conclusion is known but the path

    to that conclusion is not known, then that is reasoningbackwards.

    The more knowledge a system is given, the morecompetent it becomes.

  • 7/29/2019 expert system and decision making

    12/22

    RULE BASED ES

    Any rule consists of two parts: the IF

    part, called the antecedent(condition)

    and the THEN part called the

    consequent(action).

    IF antecedent

    THEN consequent

    The antecedent of a rule incorporates

    two parts: an objectand its value.

    A rule can have multiple antecedents

    joined by the keywords AND or OR.

  • 7/29/2019 expert system and decision making

    13/22

    INFERENCE ENGINE

    The inference engine is the genericcontrol mechanism that applies the

    knowledge present in the knowledge

    base to the task-specific data to arrive atsome conclusion.

    inference mechanisms that search

    through the database and deduce results

    in an organized manner:

    Forward chaining

    Backward chaining

    Tree searches

  • 7/29/2019 expert system and decision making

    14/22

    FORWARD AND BACKWARDCHAINING

    Forward chaining - one proceeds froma given situation toward a desired goal,adding new assertions along the way.

    Backward chaining - In this strategy,

    one starts with the desired goal, andthen attempts to find evidence forproving the goal.

    Eg. Rule 1: IF A and C THEN F

    Rule 2: IF A and E THEN GRule 3: IF B THEN ERule 4: IF G THEN D

  • 7/29/2019 expert system and decision making

    15/22

    TREE SEARCHES

    A knowledge base can usually berepresented as a branching network ortree.

    Two approaches: depth-first search. breadth-first search.

    The depth-first search algorithm begins at anode that either represents the given data

    (forward chaining) or the desired goal(backward chaining).

    Breadth-first search starts by expanding all thenodes one level below the first node. Then itsystematically expands each of these nodes till

    a solution is reached or the tree is completelyexpanded.

  • 7/29/2019 expert system and decision making

    16/22

    Tree searches (cont.)

  • 7/29/2019 expert system and decision making

    17/22

    INTELLIGENT EDITORS

    An intelligent editor acts as an interfacebetween a domain expert and an ExpertSystem.

    The editor has direct access to theknowledge in the Expert System and knows

    the structure of that knowledge Through the editors, an expert can create,

    modify and delete rules without knowledgeof the internal structure of the rules

  • 7/29/2019 expert system and decision making

    18/22

    ADVANTAGES OF EXPERTSYSTEMS

    Permanence

    human experts may forget. Reproducibility many copies. Efficiency increased system output. Consistency similar transactions give

    same output. Documentation permanent

    documentation. Completeness all transactions can be

    viewed. Timeliness Entry barriers for competitors. Differentiation

  • 7/29/2019 expert system and decision making

    19/22

    DISADVANTAGES

    Common sense

    Creativity in unusual situations.

    Learning

    no adaptability.

    Sensory Experience

    Degradation not good when noanswer exists.

  • 7/29/2019 expert system and decision making

    20/22

    APPLICATIONS

    Diagnosis and Troubleshooting ofDevices and Systems of All Kinds. Eg5GL DOCTOR

    Planning and Scheduling. Eg GATES

    Configuration of ManufacturedObjects from Subassemblies.

    Financial Decision Making. EgINSPECTOR

    Knowledge Publishing Process Monitoring and Control Design and Manufacturing. Eg

  • 7/29/2019 expert system and decision making

    21/22

    CONCLUSION

    Problem solving is accomplished byapplying specific knowledge rather than

    specific technique. This is a key idea in

    expert systems technology.

    It reflects the belief that human experts do

    not process their knowledge differently from

    others, but they do possess different

    knowledge. With this philosophy, when onefinds that their expert system does not

    produce the desired results, work begins to

    expand the knowledge base, not to

    reprogram the procedures.

  • 7/29/2019 expert system and decision making

    22/22

    THANK YOU