Fuzzy Logic and Fuzzy Cognitive Map

MATH 800 – 4
Fall 2011

Vijay Mago, Postdoctoral Fellow, The Modelling of Complex Social Systems (MoCSSy) Program, The
IRMACS Centre, Simon Fraser University, BC, Canada.
[email protected]
•  Fuzzy Logic Introduction
•  Fuzzy Numbers
•  Fuzzy Sets
•  Fuzzy Inference System
•  Examples
•  Modelling the Underground Economy in Taiwan Prof. Lotfi A. Zadeh
•  Rainfall Events Prediction
•  Fuzzy Toolbox or libraries

•  Fuzzy Cognitive Maps

•  Examples
Prof. Bart Kosko

2
Fuzzy Logic Introduction

•  Fuzzy Number

•  Number ‘x’
x-2 x-1 x x+1 x+2

•  Near ‘x’

x-2 x-1 x x+1 x+2

•  Almost ‘x’

x-2 x-1 x x+1 x+2

3
Fuzzy Logic Introduction

•  Fuzzy Sets
In a crisp set, membership or non-membership of element ‘x’ in set
A is described by a characteristic function

Fuzzy set theory extends this concept by defining partial

membership. A fuzzy set A on a universe of discourse U is
characterized by a membership function
that takes values in the interval [0, 1].

4
Fuzzy Logic Introduction

•  Fuzzy Sets...
A fuzzy set A in U may be represented as a set of ordered
pairs. Each pair consists of a generic element x and its grade of
membership function; that is

(a) Crisp membership function (b) Fuzzy membership function

5
Fuzzy Logic Introduction

•  Fuzzy Sets...
•  Fuzzy set operations

•  OR

•  AND

•  NOT

6
Fuzzy Logic Introduction

•  Fuzzy Inference System

7
Fuzzy Logic Introduction

•  Fuzzy Inference System...

Mamdani Method

•  In 1975, Professor Ebrahim Mamdani of London University built

one of the first fuzzy systems to control a steam engine and boiler
combination. He applied a set of fuzzy rules supplied by experienced
human operators.

8
Fuzzy Logic Introduction

•  Fuzzy Inference System…

9
Fuzzy Logic Introduction

•  Fuzzy Inference System…

o  An example
•  Two inputs (x, y)
•  One output (z)

•  Rules:

Rule1: If x is A3 or y is B1 Then z is C1
Rule2: If x is A2 and y is B2 Then z is C2
Rule3: If x is A1 Then z is C3

10
Fuzzy Logic Introduction

•  Fuzzy Inference System…

o  Input x: research_funding
o  Input y: project_staffing
o  Output z: risk

•  Rules:
Rule1: If research_funding is adequate or project_staffing is small Then risk is low

Rule2: If research_funding is marginal and project_staffing is large Then risk is normal

Rule3: If research_funding is inadequate Then risk is high

11
Fuzzy Logic Introduction

•  Fuzzy Inference System…

Step 1: Fuzzification

12
Fuzzy Logic Introduction

•  Fuzzy Inference System…

Step 2: Rule Evaluation

Antecedent  Consequent

13
Fuzzy Logic Introduction

•  Fuzzy Inference System…

Step 2: Rule Evaluation...

14
Fuzzy Logic Introduction

•  Fuzzy Inference System…

Step 2: Rule Evaluation...
The result of the antecedent evaluation can be applied to the
membership function of the consequent in two different ways:

15
Fuzzy Logic Introduction

•  Fuzzy Inference System…

Step 3: Rule Evaluation

16
Fuzzy Logic Introduction

•  Fuzzy Inference System…

Step 4: Defuzzification

17
Example 1:

•  A Fuzzy Logic Approach to Modeling the Underground

Economy in Taiwan

Inputs:
o  Tax Rate (TR)
o  Degree of government regulations (REG)

Output
o  The size of Underground Economy (UE)

18
Example 1…

If REG = VH and TR = VH Then UE = VB

19
Example 2:

•  Rainfall events prediction using rule-based fuzzy

inference system

Inputs:
o  Relative humidity
o  Total cloud cover
o  Wind direction
o  Temperature and
o  Surface pressure

Output
o  Rainfall events

20
Example 2…

21
Toolboxes and Libraries for FL

Fuzzy Logic Toolbox for MATLAB:

http://www.mathworks.com/products/fuzzylogic/index.html

Fuzzy Logic package for Java (jFuzzyLogic)

http://jfuzzylogic.sourceforge.net/html/index.html

Fuzzy Logic libraries for C++ (JFuzzyQt)

http://sourceforge.net/projects/jfuzzyqt/

22
Q….Q…Q ???

Q: What is fuzzy logic and why do critics call it "the cocaine

of science?"

Kosko:

Fuzzy logic is a way of doing science without math.

It's a new branch of machine intelligence that tries to make

computers think the way people think and not the
other way around.

You don't write equations for how to wash clothes.

Instead you load a chip with vague rules like “if the
wash water is dirty, add more soap,” and “if very
dirty, add a lot more.”

You can never get the science right to more than a few
decimal places. That's one reason we find chaos when
we look at things up close.
http://sipi.usc.edu/~kosko/index.html

23
Fuzzy Logic… so far

•  Over 53,000 papers listed in the INSPEC database

•  More than 15,000 in the Math Science Net database.

•  Fuzzy-logic-related patents:
•  Over 4800 in Japan
•  1500 + in the United States.

24
25
Fuzzy Cognitive Map

•  Introduction

•  Fuzzy Virtual worlds:

•  “Virtual worlds show how actors relate to one another … Events

cause one another to some degree…”

•  “Fuzzy cognitive maps (FCMs) show how causal concepts affect

one another to some degree… Causal concepts in a virtual worlds
include events, values, moods, trends, or goals…”

26
Fuzzy Cognitive Map

•  Introduction…
•  Basic structure of FCM

•  Each node in FCM represents a concept.

•  Each arc (Ci, Cj) is directed as well as weighted, and represents causal link

between concepts, showing how concept Ci causes concept Cj.

27
Fuzzy Cognitive Map

•  Introduction…
•  Basic structure of FCM…

excitatory

inhibitory

28
Fuzzy Cognitive Map

•  Introduction…
•  Basic structure of FCM…

Sanitation
-0.9/VH facilities

# of
diseases /
1000 -0.9/VH
residents

+0.8/H Bacteria
per area

A civil engineering FCM

29
Fuzzy Cognitive Map

•  Introduction…
•  Adjacency matrix

C1 C2 C3 …

C1 0 VH VL .
W= C2 H 0 0 .

C3 VL H 0 .

... . . . .

30
Fuzzy Cognitive Map

•  Introduction…
?

New state Previous state Weight matrix

31
Fuzzy Cognitive Map

•  Introduction…
•  Transfer function of FCM

(a)

(b)

(c)

32
Fuzzy Cognitive Map

•  FCM Inference Algorithm

Step 1: Definition of the initial vector A that corresponds to the elements-concepts identified by
experts’ suggestions and available knowledge.

Step 2: Multiply the initial vector A with the matrix W defined by experts

Step 3: The resultant vector A at time step k is updated using function threshold ‘ f ’.

Step 4: This new vector is considered as an initial vector in the next iteration.

Step 5: Steps 2–4 are repeated until epsilon (where epsilon is a residual, describing the minimum
error difference among the subsequent concepts)

33
Fuzzy Cognitive Map

•  Example 1:

Trivalent FCM for the control of a dolphin actor in virtual world

34
Fuzzy Cognitive Map

•  Example 2:

FCM for dolphin, fish and sharks in virtual world

35
Fuzzy Cognitive Map

•  Example 3:

FCM model for predicting the severity index of pulmonary infection

36
Fuzzy Cognitive Map

•  Example 4:

FCM differential diagnosis model of SLI from dyslexia and autism

37
FCM?

38