Programming for

Geographical Information Analysis:

Advanced Skills

Fuzzification and defuzzification

DY Patil International University,Pune

by Dr.Dipika Pradhan
 How do our statements match the
real world?

We’re assuming we know what words like “hunger” and “high”

mean in “if hunger is high go to café”.

While we might be able to give a clear RDF definition of

“hunger” it won’t ever match the real world. This may not
matter if our rules replicate reality quantitatively.

However, to do this, we need to define and use “high”.

For this, we need Fuzzy Sets and Fuzzy Logic.
DY Patil International University,Pune by
Dr.Dipika Pradhan
 Fuzziness
Traditional logic: true vs. false
Traditional empiricism: what exists or does not, what causes
something, and what doesn’t.
THESE ARE THE FOUNDATIONS OF SCIENCE.

However, since the early Greeks, people have found that the way we
understand the world isn’t black and white.

The Sorites Paradox

If I remove one sand grain from a pile, it’s still a pile, but if I carry on, it’s
soon not a pile (well, go figure…).
DY Patil International University,Pune by
Dr.Dipika Pradhan
 Reality is fuzzy
Language isn’t usually precise…
How hot is “hot”? 30 degrees? 40 degrees?

Reality isn’t actually that precise…

Most things, including us, are just continually renewing lumps of
geology! Where do we begin? Where do we end?
Atoms aren’t distinct.
Shrödinger’s poor old cat is both alive and dead.

DY Patil International University,Pune by

Dr.Dipika Pradhan
 Fuzzy Sets and Logic
.

Fuzzy Sets let us say something is 90% “one thing” and 10% “another”,
without being illogical.

Fuzzy Logic then lets us use this in rules:

E.g. it’s 90% “right” to do something, so I’ll do it 90% - adding
warm water to a washing machine cycle, for example.

DY Patil International University,Pune by

Dr.Dipika Pradhan
 The fuzzy rebellion
1923: “Bertie” Russell releases a paper on Vagueness.
How do we define objects that are partly in two normally mutually
exclusive sets?
1937: Black defines Vague sets.
1965: Zadeh coins the daft name “Fuzzy” for a logic based on Vague
set membership, instantly putting humourless scientist’s backs up.
1972: 1st practical demonstration Mamdani’s steam engine.
1981: 1st commercial applications using fuzzy logic to control
systems.
1994: Japan exported $35 billion worth of fuzzy products.
DY Patil International University,Pune by
Dr.Dipika Pradhan
 Scientists loved it so much…

“Fuzzy theory is wrong, wrong, and pernicious.

What we need is more logical thinking, not less.
The danger of fuzzy logic is that it will encourage
the sort of imprecise thinking that has brought us
so much trouble. Fuzzy logic is the cocaine of
science.”
Prof William Kahan

DY Patil International University,Pune by

Dr.Dipika Pradhan
 The advantages of Fuzzy Logic
Lets us use terms like “hot” in computers, integrating
knowledge and machine learning.

A very simple approach to building computer models of

geographical systems.

The model uses human language, so it’s reasonably

understandable.

DY Patil International University,Pune by

Dr.Dipika Pradhan
 Fuzzification
 Fuzzification is the process of converting a crisp numerical input into a
fuzzy value, represented by membership functions in a fuzzy set. It is a
key step in fuzzy logic systems, allowing them to handle uncertainty
and imprecise data.

 Suppose we want to classify temperature as cold, warm, or hot. A crisp

temperature of 28°C might not fit perfectly into a single category.
Instead, we define fuzzy sets:
 Cold: 0–20°C
 Warm: 15–30°C
 Hot: 25–40°C
 Using fuzzification, 28°C might belong to both "warm" (0.8
membership) and "hot" (0.2 membership)

DY Patil International University,Pune by

Dr.Dipika Pradhan
 Methods of Fuzzification
1. Singleton Fuzzification
 Assigns a crisp input value to a single fuzzy set with full
membership (1.0) and all other sets with zero
membership.
Example: If temperature = 30°C, it is classified only as Warm
(1.0 membership), without overlap with other fuzzy sets.
2. Triangular Fuzzification
 Uses a triangular function to represent membership.
Example: If a Speed fuzzy set has a triangular membership
function with a peak at 50 km/h and base from 30–70 km/h,
then an input of 40 km/h might have 0.5 membership in
Medium Speed.
DY Patil International University,Pune by
Dr.Dipika Pradhan
 2. Gaussian Fuzzification
 Uses a Gaussian function to define the membership of a
crisp input over a range.
Example: For a temperature input of 28°C, if the Gaussian
function for "Warm" is centered at 25°C with a standard
deviation of 5°C, the membership might be 0.8.
 3. Trapezoidal Fuzzification
 Uses a trapezoidal membership function where a crisp
input partially belongs to two or more sets.
Example: If temperature = 28°C, and "Warm" is defined from
20–35°C, with full membership between 25–30°C, then 28°C
might have 0.8 membership in Warm and 0.2 in Hot.

DY Patil International University,Pune by

Dr.Dipika Pradhan
 Defuzzification
Defuzzification is the process of converting a fuzzy
output obtained from a fuzzy logic system into a single
crisp value. This step is necessary because real-world
applications require specific numerical results rather
than fuzzy sets.

DY Patil International University,Pune by

Dr.Dipika Pradhan
Max Membership Principle
Centriod method
Weighted average method
Mean of max membership
center of sums
Largest of Maximum
Smallest of Maximum

DY Patil International University,Pune by

Dr.Dipika Pradhan
A fuzzy output is represented by a trapezoidal
membership function defined as:

Find the defuzzified value using the graphical

(integration-based centroid) method.
•a=2, b=4, c=6, d=8
•The membership function μ(x)is:

◦μ(x)=x−2/2 2≤x≤4

◦μ(x)=1 4<x<6

◦μ(x)=8−x/2 6≤x≤8

DY Patil International University,Pune by

Dr.Dipika Pradhan
 λ-cuts for Fuzzy Relations
A λ-cut (or alpha-cut) of a fuzzy relation is a crisp relation
derived from the fuzzy relation by including only those element
pairs whose membership values are greater than or equal to a
specified threshold value λ (lambda), where λ ∈ [0, 1].
Formally,
If R(x, y) is a fuzzy relation on X × Y, then the λ-cut of R is
defined as:
Rλ={(x,y)∈X×Y/μR​(x,y)≥λ}

μR(x,y) is the membership value of the pair (x, y) in the fuzzy

relation Rλis the corresponding crisp (non-fuzzy) relation at
level λ
λ-cuts help us approximate a fuzzy relation by thresholding it at
a desired level.

When λ = 0, the cut includes all non-zero membership pairs.

As λ increases, fewer pairs satisfy the condition μ R​(x,y)≥λ, hence

the relation becomes more specific.

DY Patil International University,Pune by

Dr.Dipika Pradhan
 Let’s consider a fuzzy relation R on the set X = {a, b} and Y = {1, 2, 3}
defined by the following matrix of membership values:
 𝜇𝑅=[0.9 0.6 0.2

0.4 0.7 0.1]

 This corresponds to:
 R(a,1) = 0.9, R(a,2) = 0.6, R(a,3) = 0.2
 R(b,1) = 0.4, R(b,2) = 0.7, R(b,3) = 0.1
 Let’s compute the λ-cuts:
 1. For λ = 0.5 Include pairs where
 𝜇𝑅(𝑥,𝑦)≥0.5
 R0.5={(a,1),(a,2),(b,2)}

2.For λ = 0.7
 R0.7={(𝑎,1),(𝑏,2)}

3.For λ = 0.2
 R 0.2={(a,1),(a,2),(a,3),(b,1),(b,2)}
 (Note: (b,3) is not included because its membership is 0.1 < 0.2)
DY Patil International University,Pune by
Dr.Dipika Pradhan
 Fuzzy Sets
We give things a degree of membership between 0 and 1 in several
sets (to a combined total of 1).
We then label these sets using human terms.
Encapsulates terms with no consensus definition, but we might use
surveys to define them.
Degree of membership

Membership
Young Middle Old function
0.5 Aged

0 50 90 Age
38yr old = 10% Young + 90% Middle DY Patil International University,Pu
Dr.Dipika Pradhan
 Fuzzy Logic models
We give our variables membership functions, and
express the variables as nouns (“length”,
“temperature”) or adjectives (“long”, “hot”).

We can then build up linguistic equations (“IF length

long, AND temperature hot, THEN openWindow”).

DY Patil International University,Pune by

Dr.Dipika Pradhan
Fuzzy grammar
Fuzzy grammar is a linguistic approach that extends
traditional formal grammar by incorporating the
concept of fuzziness. It is useful for handling
ambiguity, imprecise information, and natural
language processing in AI and computational
linguistics.

DY Patil International University,Pune by

Dr.Dipika Pradhan
 Fuzzy Grammar
In fuzzy grammar, each production rule is associated
with a membership value (between 0 and 1) that
represents the degree of certainty or appropriateness
of applying that rule. This is useful in natural language
processing (NLP) when dealing with ambiguity or
incomplete/informal language.

DY Patil International University,Pune by

Dr.Dipika Pradhan
1. Non-Terminals (VN):
VN = {sentence, noun phrase, verb phrase, article,
noun, verb, adverb}

2. Terminals (VT):
VT = {the, athlete, jumped, high}

DY Patil International University,Pune by

Dr.Dipika Pradhan
3. Production Rules (P) with Fuzzy Values:
Rule Production Membership
R1 sentence → noun phrase verb phrase 1.0
R2 noun phrase → article noun 0.9
R3 verb phrase → verb adverb 0.8
R4 article → the 1.0
R5 noun → athlete 1.0
R6 verb → jumped 1.0
R7 adverb → high 0.7

DY Patil International University,Pune by

Dr.Dipika Pradhan
DY Patil International University,Pune by
Dr.Dipika Pradhan
DY Patil International University,Pune by
Dr.Dipika Pradhan
Fuzzy Rule with Membership Values:

"I eat an apple." → 1.0 (Correct)

"Apple I eat." → 0.5 (Unusual but understandable)

"Eat I apple." → 0.3 (Less correct)

"Apple eat I." → 0.1 (Very incorrect)


DY
A membership function can be defined as:
Patil International University,Pune by
Dr.Dipika Pradhan
"She sings beautifully." (Correct, membership = 1.0)

"She sing beautifully." (Almost correct, membership =

0.8)

"She beautifully sings." (Somewhat correct,

membership = 0.6)

"Beautifully sings she." (Incorrect, membership = 0.3)

DY Patil International University,Pune by

Dr.Dipika Pradhan
 Crisp data How the models work
Fuzzifier Inputs converted to
Member 90% hot degrees of membership
10% cold of fuzzy sets.
Fuzzy rules
IF 90% hot THEN 80% open Fuzzy rules applied to get
IF 10% cold THEN 20% closed new sets of members.

Fuzzy output set

80% open, 20% closed These sets are then
converted back to real
Defuzzifier numbers.

Crisp data DY Patil International University,Pune by

Dr.Dipika Pradhan
 A model to work out time by foot
itchiness
Work out how bored you are (take a poll of happiness vs.
foot itchiness)…
Degree of membership

Membership
function
0.5 Happy Bored

0 5 10
Foot itchiness
0.6 bored to tears + 0.4 happy as DY Patil International University,P
Dr.Dipika Pradhan
 Decide on rules
If BORED then LECTURE LONG.
If HAPPY the LECTURE SHORT.

Make Fuzzy sets defining people’s notion of “Long” and

“Short”

Short Long

0 2
hours DY Patil International University,Pune by
 Transfer the degrees to the output sets
1
0.6 Bored
Short Long
0.4 Happy

0 2 Hours

Adjust the heights of the sets to e.g. 40% for short.

1
0.6

0.4

0 2 Hours
DY Patil International University,Pune by
Dr.Dipika Pradhan
 Calculate the average

The average is usually taken as the gravitational centre

of the sets combined.

0.6

0.4

0 2

We’re only half way through!

DY Patil International University,Pune by

Dr.Dipika Pradhan
 Uses in geography
Spatial Interaction: modelling flows between regions. Can
capture the following kind of rules..
If distance is SHORT then TRIPS are LOTS.
If distance is MEDIUM then TRIPS are some.
If distance is LONG then TRIPS are few.

GIS, remote sensing, land evaluation.

Fuzzy viewsheds.
Fuzzy GIS - expert system shell on an intelligent GIS.
Classification of land cover (vegetation, soils) from satellite imagery.

DY Patil International University,Pune by

Dr.Dipika Pradhan
 Other stuff
It’s often helpful to get other AI techniques to generate
the membership functions – e.g. Neural Nets and Genetic
Algorithms.

It’s often useful to combine several AIs with Fuzzy Logic –

e.g. you might have a Net for “low” river flows and
another for “high” ones, and a Fuzzy Logic engine
between combining the results.

DY Patil International University,Pune by

Dr.Dipika Pradhan