MYCIN example - PowerPoint PPT Presentation

1 / 7
About This Presentation
Title:

MYCIN example

Description:

'Joe passes the exam' equivalent to: Pass(Joe) need to compute measure of belief (MB) and measure of disbelief (MD) of Pass(Joe) ... – PowerPoint PPT presentation

Number of Views:254
Avg rating:3.0/5.0
Slides: 8
Provided by: pagesCpsc
Category:
Tags: mycin | example

less

Transcript and Presenter's Notes

Title: MYCIN example


1
MYCIN example
  • Joe passing

2
Description
  • If the preparation for the exam is good and the
    student slept well, then there is a good chance
    (0.7) that the student will pass the exam
  • If the students contribution to the team effort
    is high and the workload of the student is low,
    then there is a good chance (0.8) that the
    student will pass the exam.
  • If the workload of the student is high and the
    extra-curricular activities are high, then there
    is a good chance (0.6) that the student will fail
    the exam.
  • If the extra-curricular activities of the student
    are low and the health of the student is good,
    then there is a good chance (0.6) that the
    student will pass the exam.

3
Rules/Interpretations
  • (e1) Preparation(X,good) Sleep(X,well) ?1
    Pass(X) I(e1) 0.7
  • (e2) TeamContrib(X,high) Workload(X,low) ?2
    Pass(X) I(e2) 0.8
  • (e3) Workload(X,high) ExtraAct(X,high) ?3
    Pass(X)I(e3) 0.6
  • (e4) ExtraAct(X,low) Health(X,good) ?4
    Pass(X)
  • I(e4) 0.6
  • I(Preparation(Joe,good)) 0.7
  • I(Sleep(Joe,well)) 0.6
  • I(TeamContrib(Joe,high)) 0.9
  • I(Workload(Joe,low)) 0.6
  • I(ExtraAct(Joe,high)) 0.3
  • I(Health(Joe,good)) 0.8

4
Question
  • We want to interpret the statementJoe passes
    the exam
  • equivalent to Pass(Joe)
  • need to compute measure of belief (MB) and
    measure of disbelief (MD) of Pass(Joe)
  • Note range of MB and MD is 0,1 but they are
    not probabilities thus MB 1-MD (not
    necessarily)
  • range of the interpretation values I is -1,1
    where 0 is neutral, 1 is strongly agree and -1 is
    strongly disagree
  • Note because of lack of initial knowledge in the
    system, we will treat the input Is as
    probabilistic entities thus adding up to 1 for
    each object we also assume they are initially
    positive

5
Answer (MD)
  • MD(Pass(Joe),e) MD(Pass(Joe),e3) I(e3)
    max0,minI(Workload(Joe,high)),
    I(ExtraAct(Joe,high)) 0.6 max0, min1-0.6,
    0.3 0.60.3 0.18

6
Answer (MB)
  • MB(Pass(Joe),e1) I(e1) max0,minI(Prepara
    tion(Joe,good)), I(Sleep(Joe, well)) 0.7 0.6
    0.42
  • MB(Pass(Joe),e2) I(e2) max0,minI(TeamCon
    trib(Joe,high)), I(Workload(Joe, low)) 0.8
    0.6 0.48
  • MB(Pass(Joe),e4) I(e4) max0,minI(
    ExtraAct(Joe,low)), I( Health(Joe, good)) 0.6
    0.7 0.42
  • MB(Pass(Joe),e2,e4) MB(Pass(Joe),e2)
    MB(Pass(Joe),e4)(1-MB(Pass(Joe),e2))
    0.480.420.520.7
  • MB(Pass(Joe),e) MB(Pass(Joe),e1,e2,e4)
    MB(Pass(Joe),e1) MB(Pass(Joe),e2,e4)(1-MB
    (Pass(Joe),e1)) 0.420.70.580.83

7
Answer (I)
  • I(Pass(Joe)) MB(Pass(Joe)) MD(Pass(Joe))
    0.83 0.18 0.65
  • So, based on our knowledge, Joe passes the exam
    with interpretation value of 0.65.
Write a Comment
User Comments (0)
About PowerShow.com