SlideShare a Scribd company logo
1 of 42
Ordered Weighted Averaging…
Ordered Weighted Averaging
(OWA)
AmirHossein Sajedi
Master of Computer Engineering – Artificial Inteligence
Agenda
 Sensor Fusion Techniques
 Information Fusion: Mathematical Definition
 Ordered Weighted Averaging Operators
(OWA): Definition
 Learning OWA Operators:
– O’Hagan method
– Exponential OWA Operators:
• Pessimistic OWA
• Optimistic OWA
 An Application
1
Sensor Fusion Techniques
 Conventional Approaches :
OWA (Ordered Weighted Averaging) Method
Bayesian Approaches
Dempster-Shafer Method
Kalman Filter
 Intelligent (Knowledge-based) Approaches:
Fuzzy Logic (Integral Operators)
Neural Network
2
Fusion Methodology
 The most common data fusion and integration methods:
3
Fusion Function Properties
 Idempotency:
i:a ia  F(a1,…, a n) a
 Commutativity:
F(a,b)  F(b,a)
 Monotonicity:
i:a i>b i  F(a1,…, a n) > F(b1,…, b n)
4
Examples of Averaging Operators
 Mean: 𝐹 (𝑎1, 𝑎2, … , 𝑎𝑛) =
 Max
 Min
 Median
 Mode
 OWA Operators
5
OWA Operators Definition
 A mapping function 𝐹: 𝑅𝑛 → 𝑅 is called an OWA Operator of
dimension n if it has an associated weighting vector W such that:
From: D. Filev and R.R. Yager, “Learning OWA Operator Weights from Data”, In
Proceedings of the Third IEEE Conference on Fuzzy Systems, pp. 468-473, 1994.
6
OWA Operators
 Examples of OWA Operators:
7
OWA Operators
 Examples of OWA Operators:
8
Mathematical Properties of the OWA Operators
 Bounded:
 Monotonicity:
 Symmetric (or Commutativity):
 Idempotency:
9
Mathematical Properties of the OWA Operators
 Associativity:
 Stability for a linear function:
 Invariance:
10
Behavioral Properties of the OWA Operators
 Decisional behavior:
-express the behavior of the decision-maker, such as: tolerant,
optimistic,pessimistic or strict. These behaviors usually named
disjunctive and conjunctive behaviors.
 Interpretability of the parameters:
-It is to be hoped that the parameters have almost evident
semantic interpretation. This property forbids the use of a black
box methodology.
 Weights on the arguments:
-It is crucial to have the possibility to express weights on the
arguments. This can be understood as privileging some of them.
11
Measure Definition
 Orness: characterizes the degree to which the aggregation is like an
OR (Max) operation.
 Some special cases:
12
Measure Definition
 Dispersion: dispersion or entropy associated with a weighting vector.
 It was suggested for the use in calculating how much of the
information in the arguments is used during an aggregation based
on W.
13
Learning OWA Operators
 O’Hagan method: using Orness and Dispersion measures to
develop a way to generate the OWA weights while having a
predefined degree of Orness (α) and aim to maximize the entropy.
14
Learning OWA Operators From Observations
 Available information:
– a collection of m samples (observations) each comprised of an n-
tuple of values:
– aggregated value:dk
 Looking for a vector of OWA weights by minimizing the instantaneous
error ek with respect to weights wi, where:
15
Learning OWA Operators From Observations
 Constraints:
 Transformation: we shall express the OWA weights as follows
16
Unconstrained Minimization Problem
 Minimize the instantaneous error 𝑒 𝑘 with respect to the parameters 𝜆𝑖:
 Using the gradient descent method, we obtain the following rule for updating
the parameters 𝜆𝑖:
 where β denotes the Learning Rate (0≤ β ≤ 1)
17
Unconstrained Minimization Problem
 Therefore, the parameters 𝜆𝑖 determining the OWA
weights, are updated by back-propagation of the error (𝑑^ 𝑘 − 𝑑 𝑘 )
between the current estimated aggregated value and the actual aggregated
value with factors 𝑤𝑖 and (𝑏 𝑘𝑖−𝑑^ 𝑘).
 These factors are the current OWA weight 𝑤𝑖 and the difference (𝑏 𝑘𝑖
−𝑑^ 𝑘) between the i-th aggregate object 𝑏𝑘𝑖 and the current
estimated aggregated 𝑑^ 𝑘 .
18
Learning OWA Weights: Example 1
19
Unconstrained Minimization Problem
 Applying the learning algorithm:
– Initial values:
– Learning rate:
– After 150 iterations:
20
Learning OWA Weights: Example 2
21
Learning OWA Weights: Example 2
 Applying the learning algorithm:
– Initial values:
– Learning rate:
– After 50 iterations:
22
Exponential OWA Operators
 O’Hagan method: involves the solution of a constrainednonlinear
programming problem
 Exponential OWA Operators: very simple relationship exists between
the Orness function and the parameter which determines the OWA
weights.
 Usefulness: Generation of OWA weights satisfying a given degree of
Orness.
23
Optimistic Exponential OWA Operators
 Definition (Optimistic Exponential OWA Operators):
where parameter α belongs to the unit interval (0≤ α ≤ 1)
 A recursive express of these weights is as:
24
Optimistic Exponential OWA Operators
 An alternative view to the process of generation of the Optimistic
Exponential OWA weights:
 Assume W is a weighting vector of dimension n, we extend it to a
vector V of dimension n+1 as follows:
 Thus in extending this, we simply proportionally divide the weight in
the last place of the old vector.
25
Optimistic Exponential OWA Operators
 Properties:
 Thus, from a formal point of view these weights can serve as OWA
weights.
 Special cases:
26
Optimistic Exponential OWA Operators
 Orness of the Optimistic Exponential OWA Operators:
27
Optimistic Exponential OWA Operators
 For the Optimistic Exponential OWA Operators, the Orness function
is a monotonically increasing function of parameter α.
 As the number of arguments increases, this aggregation becomes
more and more OR-like.
28
Optimistic Exponential OWA Operators
 The functional relationship between the Orness and
parameter α for different numbers of arguments
n=2.3.. 10.
 The monotonically increasing nature of this
relationship with respect to α, for a fixed n, the Orness
of this aggregation increases as α increases.
 The monotonically increasing nature of this
relationship with respect to n, for a fixed α, the Orness
of this aggregation increases as n increases.
 For n=2 the Orness value of this aggregation is
always equal to α.
 For any value of n>2, the degree of Orness is higher
than the value of the parameter α. 32
29
Pessimistic Exponential OWA Operators
 Definition:
where parameter α belongs to the unit interval (0≤ α ≤ 1)
 A recursive express of these weights is as:
30
Pessimistic Exponential OWA Operators
 An alternative view to the process of generation of the Pessimistic
Exponential OWA weights:
 Assume W is a weighting vector of dimension n, we extend it to a
vector V of dimension n+1 as follows:
31
Pessimistic Exponential OWA Operators
 Properties:
 Thus, from a formal point of view these weights can serve as OWA
weights.
 Special cases:
32
Pessimistic Exponential OWA Operators
 For the Pessimistic Exponential OWA Operators, the Orness unction
is a monotonically increasing function of parameter α.
 Orness function is a monotonically decreasing function of paramete n.
 As the number of arguments increases, this aggregation becomes
more and more AND-like.
33
Pessimistic Exponential OWA Operators
 The functional relationship between the Orness and
parameter α for different numbers of arguments n = 2.
3 .... 10.
 The monotonically increasing nature of this
relationship with respect to α, for a fixed n, the Orness
of this aggregation increases as α increases.
 The monotonically decreasing nature of this
relationship with respect to n, for a fixed α, the Orness
of this aggregation decreases as n increases.
 For n=2 the Orness value of this aggregation is
always equal to α. • For any value of n>2, the degree
of Orness is lower than the value of the parameter α.
34
Exponential OWA Operators
 The Optimistic and Pessimistic Exponential OWA Operators have one
very useful property:
– Given a value of n and a desired degree of Orness, one can simply
obtain from previous figures the associated value of α. Then,
the OWA weights can easily be generated according to the
corresponding equations.
 This simple technique eliminates the need to calculate the OWA
weights via the sophisticated O'Hagan’s procedure, which requires
the solution of the constrained optimization problem.
35
Exponential OWA Operators
 Let us assume the number of aggregate objects is 5. Also, assume
the desired degree of Orness is 0.6.
36
Exponential OWA Operators: An Example
 Let us assume the number of aggregate objects is 5. Also, assume
the desired degree of Orness is 0.9.
37
An Application:Result Fusion in Multi-Agent Systems
 In multi-agent systems, each of the agents may have its own expertise.When
they are asked to accomplish the same task, the results may be different.
 In such situations, it is required to fuse the results to obtain a final one.
 This paper [Z. Zhang and C. Zhang, “Result Fusion in Multi-Agent Systems
Based on OWA Operator”, 23rd Australasian Computer Science Conference
(ACSC 2000), 2000]:
– proposes an information fusion approach when multiple agents are
asked to collect the information for the same query from different
resources.
– Discusses about the group decision fusion problem that several
agents with similar knowledge make decisions using the same
information.
38
An Application:Result Fusion in Multi-Agent Systems
 Application domain:
– Financial Investment Planner Using Intelligent Agent Technologies
 Sub-systems:
– User interface and result visualization agents
– Decision-making agents
– Information retrieval and information filter agents
– Data mining agents
39
Framework of Financial Investment Planner
37
Thank you

More Related Content

What's hot

Distribusi normal, f,t
Distribusi normal, f,tDistribusi normal, f,t
Distribusi normal, f,tratuilma
 
Matematika Diskrit part 1
Matematika Diskrit part 1Matematika Diskrit part 1
Matematika Diskrit part 1radar radius
 
Metode simpleks dua fase
Metode simpleks dua faseMetode simpleks dua fase
Metode simpleks dua fasespecy1234
 
Metode interpolasi linier
Metode  interpolasi linierMetode  interpolasi linier
Metode interpolasi linierokti agung
 
Metode numerik pada persamaan integral (new)
Metode numerik pada persamaan integral (new)Metode numerik pada persamaan integral (new)
Metode numerik pada persamaan integral (new)Khubab Basari
 
Metode numerik pertemuan 7 (interpolasi lagrange)
Metode numerik pertemuan 7 (interpolasi lagrange)Metode numerik pertemuan 7 (interpolasi lagrange)
Metode numerik pertemuan 7 (interpolasi lagrange)Nerossi Jonathan
 
02 first order differential equations
02 first order differential equations02 first order differential equations
02 first order differential equationsvansi007
 
03. matrik-dan-transformasi-linear-ortonormal-dan-gram-schmidt
03. matrik-dan-transformasi-linear-ortonormal-dan-gram-schmidt03. matrik-dan-transformasi-linear-ortonormal-dan-gram-schmidt
03. matrik-dan-transformasi-linear-ortonormal-dan-gram-schmidtkmaguswira
 
Fourier analysis of signals and systems
Fourier analysis of signals and systemsFourier analysis of signals and systems
Fourier analysis of signals and systemsBabul Islam
 
Materi Barisan Matematika
Materi Barisan MatematikaMateri Barisan Matematika
Materi Barisan MatematikaHafsa RI
 
Makalah interpolasi kelompok 2
Makalah interpolasi kelompok 2Makalah interpolasi kelompok 2
Makalah interpolasi kelompok 2Arin Ayundhita
 
Deret taylor and mac laurin
Deret taylor and mac laurinDeret taylor and mac laurin
Deret taylor and mac laurinMoch Hasanudin
 
INTEGRAL menggunakan MAPLE
INTEGRAL menggunakan MAPLEINTEGRAL menggunakan MAPLE
INTEGRAL menggunakan MAPLEDyas Arientiyya
 
Game Making in Alice - Coding Preview: "Eat the Banana" Game
Game Making in Alice - Coding Preview: "Eat the Banana" GameGame Making in Alice - Coding Preview: "Eat the Banana" Game
Game Making in Alice - Coding Preview: "Eat the Banana" GameEmma M. Byrd
 
09 a analis_vektor
09 a analis_vektor09 a analis_vektor
09 a analis_vektorTri Wahyuni
 
Interpolasi lagrange dan newton
Interpolasi lagrange dan newtonInterpolasi lagrange dan newton
Interpolasi lagrange dan newtonYuni Dwi Utami
 
04 deret-fourier-gt
04 deret-fourier-gt04 deret-fourier-gt
04 deret-fourier-gtLukman Hakim
 

What's hot (20)

Distribusi normal, f,t
Distribusi normal, f,tDistribusi normal, f,t
Distribusi normal, f,t
 
Matematika Diskrit part 1
Matematika Diskrit part 1Matematika Diskrit part 1
Matematika Diskrit part 1
 
Metode simpleks dua fase
Metode simpleks dua faseMetode simpleks dua fase
Metode simpleks dua fase
 
Metode interpolasi linier
Metode  interpolasi linierMetode  interpolasi linier
Metode interpolasi linier
 
Metode numerik pada persamaan integral (new)
Metode numerik pada persamaan integral (new)Metode numerik pada persamaan integral (new)
Metode numerik pada persamaan integral (new)
 
Metode numerik pertemuan 7 (interpolasi lagrange)
Metode numerik pertemuan 7 (interpolasi lagrange)Metode numerik pertemuan 7 (interpolasi lagrange)
Metode numerik pertemuan 7 (interpolasi lagrange)
 
02 first order differential equations
02 first order differential equations02 first order differential equations
02 first order differential equations
 
03. matrik-dan-transformasi-linear-ortonormal-dan-gram-schmidt
03. matrik-dan-transformasi-linear-ortonormal-dan-gram-schmidt03. matrik-dan-transformasi-linear-ortonormal-dan-gram-schmidt
03. matrik-dan-transformasi-linear-ortonormal-dan-gram-schmidt
 
Interpolasi linier
Interpolasi linierInterpolasi linier
Interpolasi linier
 
Fourier analysis of signals and systems
Fourier analysis of signals and systemsFourier analysis of signals and systems
Fourier analysis of signals and systems
 
Irisan 2 lingkaran
Irisan 2 lingkaranIrisan 2 lingkaran
Irisan 2 lingkaran
 
Materi Barisan Matematika
Materi Barisan MatematikaMateri Barisan Matematika
Materi Barisan Matematika
 
Makalah interpolasi kelompok 2
Makalah interpolasi kelompok 2Makalah interpolasi kelompok 2
Makalah interpolasi kelompok 2
 
Deret taylor and mac laurin
Deret taylor and mac laurinDeret taylor and mac laurin
Deret taylor and mac laurin
 
INTEGRAL menggunakan MAPLE
INTEGRAL menggunakan MAPLEINTEGRAL menggunakan MAPLE
INTEGRAL menggunakan MAPLE
 
Game Making in Alice - Coding Preview: "Eat the Banana" Game
Game Making in Alice - Coding Preview: "Eat the Banana" GameGame Making in Alice - Coding Preview: "Eat the Banana" Game
Game Making in Alice - Coding Preview: "Eat the Banana" Game
 
09 a analis_vektor
09 a analis_vektor09 a analis_vektor
09 a analis_vektor
 
Interpolasi lagrange dan newton
Interpolasi lagrange dan newtonInterpolasi lagrange dan newton
Interpolasi lagrange dan newton
 
04 deret-fourier-gt
04 deret-fourier-gt04 deret-fourier-gt
04 deret-fourier-gt
 
Peubah acak-diskret-khusus
Peubah acak-diskret-khususPeubah acak-diskret-khusus
Peubah acak-diskret-khusus
 

Similar to Owa method

Numerical Methods - Power Method for Eigen values
Numerical Methods - Power Method for Eigen valuesNumerical Methods - Power Method for Eigen values
Numerical Methods - Power Method for Eigen valuesDr. Nirav Vyas
 
ML_Unit_2_Part_A
ML_Unit_2_Part_AML_Unit_2_Part_A
ML_Unit_2_Part_ASrimatre K
 
Personalized list recommendation based on multi armed bandit algorithms
Personalized list recommendation based on multi armed bandit algorithmsPersonalized list recommendation based on multi armed bandit algorithms
Personalized list recommendation based on multi armed bandit algorithmsWeiwen Liu
 
Artificial Neural Networks 1
Artificial Neural Networks 1Artificial Neural Networks 1
Artificial Neural Networks 1swapnac12
 
PPTChapter12.pdf
PPTChapter12.pdfPPTChapter12.pdf
PPTChapter12.pdfVara Prasad
 
Learning Algorithms For A Specific Configuration Of The Quantron
Learning Algorithms For A Specific Configuration Of The QuantronLearning Algorithms For A Specific Configuration Of The Quantron
Learning Algorithms For A Specific Configuration Of The Quantronsdemontigny
 
ARTIFICIAL-NEURAL-NETWORKMACHINELEARNING
ARTIFICIAL-NEURAL-NETWORKMACHINELEARNINGARTIFICIAL-NEURAL-NETWORKMACHINELEARNING
ARTIFICIAL-NEURAL-NETWORKMACHINELEARNINGmohanapriyastp
 
PRML Chapter 4
PRML Chapter 4PRML Chapter 4
PRML Chapter 4Sunwoo Kim
 
Machine Learning Algorithms Review(Part 2)
Machine Learning Algorithms Review(Part 2)Machine Learning Algorithms Review(Part 2)
Machine Learning Algorithms Review(Part 2)Zihui Li
 
Principles of soft computing-Associative memory networks
Principles of soft computing-Associative memory networksPrinciples of soft computing-Associative memory networks
Principles of soft computing-Associative memory networksSivagowry Shathesh
 
Computational electromagnetics
Computational electromagneticsComputational electromagnetics
Computational electromagneticsAwaab Fakih
 
Artificial Neural Network by Dr.C.R.Dhivyaa Kongu Engineering College
Artificial Neural Network by Dr.C.R.Dhivyaa Kongu Engineering CollegeArtificial Neural Network by Dr.C.R.Dhivyaa Kongu Engineering College
Artificial Neural Network by Dr.C.R.Dhivyaa Kongu Engineering CollegeDhivyaa C.R
 
Machine learning Module-2, 6th Semester Elective
Machine learning Module-2, 6th Semester ElectiveMachine learning Module-2, 6th Semester Elective
Machine learning Module-2, 6th Semester ElectiveMayuraD1
 
Perceptron Study Material with XOR example
Perceptron Study Material with XOR examplePerceptron Study Material with XOR example
Perceptron Study Material with XOR exampleGSURESHKUMAR11
 

Similar to Owa method (20)

Numerical Methods - Power Method for Eigen values
Numerical Methods - Power Method for Eigen valuesNumerical Methods - Power Method for Eigen values
Numerical Methods - Power Method for Eigen values
 
ML_Unit_2_Part_A
ML_Unit_2_Part_AML_Unit_2_Part_A
ML_Unit_2_Part_A
 
Personalized list recommendation based on multi armed bandit algorithms
Personalized list recommendation based on multi armed bandit algorithmsPersonalized list recommendation based on multi armed bandit algorithms
Personalized list recommendation based on multi armed bandit algorithms
 
Artificial Neural Networks 1
Artificial Neural Networks 1Artificial Neural Networks 1
Artificial Neural Networks 1
 
PPTChapter12.pdf
PPTChapter12.pdfPPTChapter12.pdf
PPTChapter12.pdf
 
Learning Algorithms For A Specific Configuration Of The Quantron
Learning Algorithms For A Specific Configuration Of The QuantronLearning Algorithms For A Specific Configuration Of The Quantron
Learning Algorithms For A Specific Configuration Of The Quantron
 
ARTIFICIAL-NEURAL-NETWORKMACHINELEARNING
ARTIFICIAL-NEURAL-NETWORKMACHINELEARNINGARTIFICIAL-NEURAL-NETWORKMACHINELEARNING
ARTIFICIAL-NEURAL-NETWORKMACHINELEARNING
 
PRML Chapter 4
PRML Chapter 4PRML Chapter 4
PRML Chapter 4
 
Machine Learning Algorithms Review(Part 2)
Machine Learning Algorithms Review(Part 2)Machine Learning Algorithms Review(Part 2)
Machine Learning Algorithms Review(Part 2)
 
Principles of soft computing-Associative memory networks
Principles of soft computing-Associative memory networksPrinciples of soft computing-Associative memory networks
Principles of soft computing-Associative memory networks
 
Computational electromagnetics
Computational electromagneticsComputational electromagnetics
Computational electromagnetics
 
Generative models
Generative modelsGenerative models
Generative models
 
UNIT III (8).pptx
UNIT III (8).pptxUNIT III (8).pptx
UNIT III (8).pptx
 
Artificial Neural Network by Dr.C.R.Dhivyaa Kongu Engineering College
Artificial Neural Network by Dr.C.R.Dhivyaa Kongu Engineering CollegeArtificial Neural Network by Dr.C.R.Dhivyaa Kongu Engineering College
Artificial Neural Network by Dr.C.R.Dhivyaa Kongu Engineering College
 
Machine learning Module-2, 6th Semester Elective
Machine learning Module-2, 6th Semester ElectiveMachine learning Module-2, 6th Semester Elective
Machine learning Module-2, 6th Semester Elective
 
Regularization
RegularizationRegularization
Regularization
 
Efficient projections
Efficient projectionsEfficient projections
Efficient projections
 
Efficient projections
Efficient projectionsEfficient projections
Efficient projections
 
Perceptron Study Material with XOR example
Perceptron Study Material with XOR examplePerceptron Study Material with XOR example
Perceptron Study Material with XOR example
 
PROJECT
PROJECTPROJECT
PROJECT
 

Recently uploaded

STERILITY TESTING OF PHARMACEUTICALS ppt by DR.C.P.PRINCE
STERILITY TESTING OF PHARMACEUTICALS ppt by DR.C.P.PRINCESTERILITY TESTING OF PHARMACEUTICALS ppt by DR.C.P.PRINCE
STERILITY TESTING OF PHARMACEUTICALS ppt by DR.C.P.PRINCEPRINCE C P
 
All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...
All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...
All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...Sérgio Sacani
 
Animal Communication- Auditory and Visual.pptx
Animal Communication- Auditory and Visual.pptxAnimal Communication- Auditory and Visual.pptx
Animal Communication- Auditory and Visual.pptxUmerFayaz5
 
Labelling Requirements and Label Claims for Dietary Supplements and Recommend...
Labelling Requirements and Label Claims for Dietary Supplements and Recommend...Labelling Requirements and Label Claims for Dietary Supplements and Recommend...
Labelling Requirements and Label Claims for Dietary Supplements and Recommend...Lokesh Kothari
 
PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...
PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...
PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...Sérgio Sacani
 
Discovery of an Accretion Streamer and a Slow Wide-angle Outflow around FUOri...
Discovery of an Accretion Streamer and a Slow Wide-angle Outflow around FUOri...Discovery of an Accretion Streamer and a Slow Wide-angle Outflow around FUOri...
Discovery of an Accretion Streamer and a Slow Wide-angle Outflow around FUOri...Sérgio Sacani
 
Isotopic evidence of long-lived volcanism on Io
Isotopic evidence of long-lived volcanism on IoIsotopic evidence of long-lived volcanism on Io
Isotopic evidence of long-lived volcanism on IoSérgio Sacani
 
G9 Science Q4- Week 1-2 Projectile Motion.ppt
G9 Science Q4- Week 1-2 Projectile Motion.pptG9 Science Q4- Week 1-2 Projectile Motion.ppt
G9 Science Q4- Week 1-2 Projectile Motion.pptMAESTRELLAMesa2
 
Nightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43b
Nightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43bNightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43b
Nightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43bSérgio Sacani
 
Analytical Profile of Coleus Forskohlii | Forskolin .pdf
Analytical Profile of Coleus Forskohlii | Forskolin .pdfAnalytical Profile of Coleus Forskohlii | Forskolin .pdf
Analytical Profile of Coleus Forskohlii | Forskolin .pdfSwapnil Therkar
 
Stunning ➥8448380779▻ Call Girls In Panchshil Enclave Delhi NCR
Stunning ➥8448380779▻ Call Girls In Panchshil Enclave Delhi NCRStunning ➥8448380779▻ Call Girls In Panchshil Enclave Delhi NCR
Stunning ➥8448380779▻ Call Girls In Panchshil Enclave Delhi NCRDelhi Call girls
 
zoogeography of pakistan.pptx fauna of Pakistan
zoogeography of pakistan.pptx fauna of Pakistanzoogeography of pakistan.pptx fauna of Pakistan
zoogeography of pakistan.pptx fauna of Pakistanzohaibmir069
 
Biological Classification BioHack (3).pdf
Biological Classification BioHack (3).pdfBiological Classification BioHack (3).pdf
Biological Classification BioHack (3).pdfmuntazimhurra
 
Recombinant DNA technology (Immunological screening)
Recombinant DNA technology (Immunological screening)Recombinant DNA technology (Immunological screening)
Recombinant DNA technology (Immunological screening)PraveenaKalaiselvan1
 
Luciferase in rDNA technology (biotechnology).pptx
Luciferase in rDNA technology (biotechnology).pptxLuciferase in rDNA technology (biotechnology).pptx
Luciferase in rDNA technology (biotechnology).pptxAleenaTreesaSaji
 
Cultivation of KODO MILLET . made by Ghanshyam pptx
Cultivation of KODO MILLET . made by Ghanshyam pptxCultivation of KODO MILLET . made by Ghanshyam pptx
Cultivation of KODO MILLET . made by Ghanshyam pptxpradhanghanshyam7136
 
Call Us ≽ 9953322196 ≼ Call Girls In Mukherjee Nagar(Delhi) |
Call Us ≽ 9953322196 ≼ Call Girls In Mukherjee Nagar(Delhi) |Call Us ≽ 9953322196 ≼ Call Girls In Mukherjee Nagar(Delhi) |
Call Us ≽ 9953322196 ≼ Call Girls In Mukherjee Nagar(Delhi) |aasikanpl
 
Biopesticide (2).pptx .This slides helps to know the different types of biop...
Biopesticide (2).pptx  .This slides helps to know the different types of biop...Biopesticide (2).pptx  .This slides helps to know the different types of biop...
Biopesticide (2).pptx .This slides helps to know the different types of biop...RohitNehra6
 
Scheme-of-Work-Science-Stage-4 cambridge science.docx
Scheme-of-Work-Science-Stage-4 cambridge science.docxScheme-of-Work-Science-Stage-4 cambridge science.docx
Scheme-of-Work-Science-Stage-4 cambridge science.docxyaramohamed343013
 

Recently uploaded (20)

STERILITY TESTING OF PHARMACEUTICALS ppt by DR.C.P.PRINCE
STERILITY TESTING OF PHARMACEUTICALS ppt by DR.C.P.PRINCESTERILITY TESTING OF PHARMACEUTICALS ppt by DR.C.P.PRINCE
STERILITY TESTING OF PHARMACEUTICALS ppt by DR.C.P.PRINCE
 
All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...
All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...
All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...
 
Animal Communication- Auditory and Visual.pptx
Animal Communication- Auditory and Visual.pptxAnimal Communication- Auditory and Visual.pptx
Animal Communication- Auditory and Visual.pptx
 
Labelling Requirements and Label Claims for Dietary Supplements and Recommend...
Labelling Requirements and Label Claims for Dietary Supplements and Recommend...Labelling Requirements and Label Claims for Dietary Supplements and Recommend...
Labelling Requirements and Label Claims for Dietary Supplements and Recommend...
 
PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...
PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...
PossibleEoarcheanRecordsoftheGeomagneticFieldPreservedintheIsuaSupracrustalBe...
 
Discovery of an Accretion Streamer and a Slow Wide-angle Outflow around FUOri...
Discovery of an Accretion Streamer and a Slow Wide-angle Outflow around FUOri...Discovery of an Accretion Streamer and a Slow Wide-angle Outflow around FUOri...
Discovery of an Accretion Streamer and a Slow Wide-angle Outflow around FUOri...
 
Isotopic evidence of long-lived volcanism on Io
Isotopic evidence of long-lived volcanism on IoIsotopic evidence of long-lived volcanism on Io
Isotopic evidence of long-lived volcanism on Io
 
G9 Science Q4- Week 1-2 Projectile Motion.ppt
G9 Science Q4- Week 1-2 Projectile Motion.pptG9 Science Q4- Week 1-2 Projectile Motion.ppt
G9 Science Q4- Week 1-2 Projectile Motion.ppt
 
Nightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43b
Nightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43bNightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43b
Nightside clouds and disequilibrium chemistry on the hot Jupiter WASP-43b
 
Analytical Profile of Coleus Forskohlii | Forskolin .pdf
Analytical Profile of Coleus Forskohlii | Forskolin .pdfAnalytical Profile of Coleus Forskohlii | Forskolin .pdf
Analytical Profile of Coleus Forskohlii | Forskolin .pdf
 
Stunning ➥8448380779▻ Call Girls In Panchshil Enclave Delhi NCR
Stunning ➥8448380779▻ Call Girls In Panchshil Enclave Delhi NCRStunning ➥8448380779▻ Call Girls In Panchshil Enclave Delhi NCR
Stunning ➥8448380779▻ Call Girls In Panchshil Enclave Delhi NCR
 
zoogeography of pakistan.pptx fauna of Pakistan
zoogeography of pakistan.pptx fauna of Pakistanzoogeography of pakistan.pptx fauna of Pakistan
zoogeography of pakistan.pptx fauna of Pakistan
 
Biological Classification BioHack (3).pdf
Biological Classification BioHack (3).pdfBiological Classification BioHack (3).pdf
Biological Classification BioHack (3).pdf
 
Recombinant DNA technology (Immunological screening)
Recombinant DNA technology (Immunological screening)Recombinant DNA technology (Immunological screening)
Recombinant DNA technology (Immunological screening)
 
Luciferase in rDNA technology (biotechnology).pptx
Luciferase in rDNA technology (biotechnology).pptxLuciferase in rDNA technology (biotechnology).pptx
Luciferase in rDNA technology (biotechnology).pptx
 
Cultivation of KODO MILLET . made by Ghanshyam pptx
Cultivation of KODO MILLET . made by Ghanshyam pptxCultivation of KODO MILLET . made by Ghanshyam pptx
Cultivation of KODO MILLET . made by Ghanshyam pptx
 
The Philosophy of Science
The Philosophy of ScienceThe Philosophy of Science
The Philosophy of Science
 
Call Us ≽ 9953322196 ≼ Call Girls In Mukherjee Nagar(Delhi) |
Call Us ≽ 9953322196 ≼ Call Girls In Mukherjee Nagar(Delhi) |Call Us ≽ 9953322196 ≼ Call Girls In Mukherjee Nagar(Delhi) |
Call Us ≽ 9953322196 ≼ Call Girls In Mukherjee Nagar(Delhi) |
 
Biopesticide (2).pptx .This slides helps to know the different types of biop...
Biopesticide (2).pptx  .This slides helps to know the different types of biop...Biopesticide (2).pptx  .This slides helps to know the different types of biop...
Biopesticide (2).pptx .This slides helps to know the different types of biop...
 
Scheme-of-Work-Science-Stage-4 cambridge science.docx
Scheme-of-Work-Science-Stage-4 cambridge science.docxScheme-of-Work-Science-Stage-4 cambridge science.docx
Scheme-of-Work-Science-Stage-4 cambridge science.docx
 

Owa method

  • 1. Ordered Weighted Averaging… Ordered Weighted Averaging (OWA) AmirHossein Sajedi Master of Computer Engineering – Artificial Inteligence
  • 2. Agenda  Sensor Fusion Techniques  Information Fusion: Mathematical Definition  Ordered Weighted Averaging Operators (OWA): Definition  Learning OWA Operators: – O’Hagan method – Exponential OWA Operators: • Pessimistic OWA • Optimistic OWA  An Application 1
  • 3. Sensor Fusion Techniques  Conventional Approaches : OWA (Ordered Weighted Averaging) Method Bayesian Approaches Dempster-Shafer Method Kalman Filter  Intelligent (Knowledge-based) Approaches: Fuzzy Logic (Integral Operators) Neural Network 2
  • 4. Fusion Methodology  The most common data fusion and integration methods: 3
  • 5. Fusion Function Properties  Idempotency: i:a ia  F(a1,…, a n) a  Commutativity: F(a,b)  F(b,a)  Monotonicity: i:a i>b i  F(a1,…, a n) > F(b1,…, b n) 4
  • 6. Examples of Averaging Operators  Mean: 𝐹 (𝑎1, 𝑎2, … , 𝑎𝑛) =  Max  Min  Median  Mode  OWA Operators 5
  • 7. OWA Operators Definition  A mapping function 𝐹: 𝑅𝑛 → 𝑅 is called an OWA Operator of dimension n if it has an associated weighting vector W such that: From: D. Filev and R.R. Yager, “Learning OWA Operator Weights from Data”, In Proceedings of the Third IEEE Conference on Fuzzy Systems, pp. 468-473, 1994. 6
  • 8. OWA Operators  Examples of OWA Operators: 7
  • 9. OWA Operators  Examples of OWA Operators: 8
  • 10. Mathematical Properties of the OWA Operators  Bounded:  Monotonicity:  Symmetric (or Commutativity):  Idempotency: 9
  • 11. Mathematical Properties of the OWA Operators  Associativity:  Stability for a linear function:  Invariance: 10
  • 12. Behavioral Properties of the OWA Operators  Decisional behavior: -express the behavior of the decision-maker, such as: tolerant, optimistic,pessimistic or strict. These behaviors usually named disjunctive and conjunctive behaviors.  Interpretability of the parameters: -It is to be hoped that the parameters have almost evident semantic interpretation. This property forbids the use of a black box methodology.  Weights on the arguments: -It is crucial to have the possibility to express weights on the arguments. This can be understood as privileging some of them. 11
  • 13. Measure Definition  Orness: characterizes the degree to which the aggregation is like an OR (Max) operation.  Some special cases: 12
  • 14. Measure Definition  Dispersion: dispersion or entropy associated with a weighting vector.  It was suggested for the use in calculating how much of the information in the arguments is used during an aggregation based on W. 13
  • 15. Learning OWA Operators  O’Hagan method: using Orness and Dispersion measures to develop a way to generate the OWA weights while having a predefined degree of Orness (α) and aim to maximize the entropy. 14
  • 16. Learning OWA Operators From Observations  Available information: – a collection of m samples (observations) each comprised of an n- tuple of values: – aggregated value:dk  Looking for a vector of OWA weights by minimizing the instantaneous error ek with respect to weights wi, where: 15
  • 17. Learning OWA Operators From Observations  Constraints:  Transformation: we shall express the OWA weights as follows 16
  • 18. Unconstrained Minimization Problem  Minimize the instantaneous error 𝑒 𝑘 with respect to the parameters 𝜆𝑖:  Using the gradient descent method, we obtain the following rule for updating the parameters 𝜆𝑖:  where β denotes the Learning Rate (0≤ β ≤ 1) 17
  • 19. Unconstrained Minimization Problem  Therefore, the parameters 𝜆𝑖 determining the OWA weights, are updated by back-propagation of the error (𝑑^ 𝑘 − 𝑑 𝑘 ) between the current estimated aggregated value and the actual aggregated value with factors 𝑤𝑖 and (𝑏 𝑘𝑖−𝑑^ 𝑘).  These factors are the current OWA weight 𝑤𝑖 and the difference (𝑏 𝑘𝑖 −𝑑^ 𝑘) between the i-th aggregate object 𝑏𝑘𝑖 and the current estimated aggregated 𝑑^ 𝑘 . 18
  • 20. Learning OWA Weights: Example 1 19
  • 21. Unconstrained Minimization Problem  Applying the learning algorithm: – Initial values: – Learning rate: – After 150 iterations: 20
  • 22. Learning OWA Weights: Example 2 21
  • 23. Learning OWA Weights: Example 2  Applying the learning algorithm: – Initial values: – Learning rate: – After 50 iterations: 22
  • 24. Exponential OWA Operators  O’Hagan method: involves the solution of a constrainednonlinear programming problem  Exponential OWA Operators: very simple relationship exists between the Orness function and the parameter which determines the OWA weights.  Usefulness: Generation of OWA weights satisfying a given degree of Orness. 23
  • 25. Optimistic Exponential OWA Operators  Definition (Optimistic Exponential OWA Operators): where parameter α belongs to the unit interval (0≤ α ≤ 1)  A recursive express of these weights is as: 24
  • 26. Optimistic Exponential OWA Operators  An alternative view to the process of generation of the Optimistic Exponential OWA weights:  Assume W is a weighting vector of dimension n, we extend it to a vector V of dimension n+1 as follows:  Thus in extending this, we simply proportionally divide the weight in the last place of the old vector. 25
  • 27. Optimistic Exponential OWA Operators  Properties:  Thus, from a formal point of view these weights can serve as OWA weights.  Special cases: 26
  • 28. Optimistic Exponential OWA Operators  Orness of the Optimistic Exponential OWA Operators: 27
  • 29. Optimistic Exponential OWA Operators  For the Optimistic Exponential OWA Operators, the Orness function is a monotonically increasing function of parameter α.  As the number of arguments increases, this aggregation becomes more and more OR-like. 28
  • 30. Optimistic Exponential OWA Operators  The functional relationship between the Orness and parameter α for different numbers of arguments n=2.3.. 10.  The monotonically increasing nature of this relationship with respect to α, for a fixed n, the Orness of this aggregation increases as α increases.  The monotonically increasing nature of this relationship with respect to n, for a fixed α, the Orness of this aggregation increases as n increases.  For n=2 the Orness value of this aggregation is always equal to α.  For any value of n>2, the degree of Orness is higher than the value of the parameter α. 32 29
  • 31. Pessimistic Exponential OWA Operators  Definition: where parameter α belongs to the unit interval (0≤ α ≤ 1)  A recursive express of these weights is as: 30
  • 32. Pessimistic Exponential OWA Operators  An alternative view to the process of generation of the Pessimistic Exponential OWA weights:  Assume W is a weighting vector of dimension n, we extend it to a vector V of dimension n+1 as follows: 31
  • 33. Pessimistic Exponential OWA Operators  Properties:  Thus, from a formal point of view these weights can serve as OWA weights.  Special cases: 32
  • 34. Pessimistic Exponential OWA Operators  For the Pessimistic Exponential OWA Operators, the Orness unction is a monotonically increasing function of parameter α.  Orness function is a monotonically decreasing function of paramete n.  As the number of arguments increases, this aggregation becomes more and more AND-like. 33
  • 35. Pessimistic Exponential OWA Operators  The functional relationship between the Orness and parameter α for different numbers of arguments n = 2. 3 .... 10.  The monotonically increasing nature of this relationship with respect to α, for a fixed n, the Orness of this aggregation increases as α increases.  The monotonically decreasing nature of this relationship with respect to n, for a fixed α, the Orness of this aggregation decreases as n increases.  For n=2 the Orness value of this aggregation is always equal to α. • For any value of n>2, the degree of Orness is lower than the value of the parameter α. 34
  • 36. Exponential OWA Operators  The Optimistic and Pessimistic Exponential OWA Operators have one very useful property: – Given a value of n and a desired degree of Orness, one can simply obtain from previous figures the associated value of α. Then, the OWA weights can easily be generated according to the corresponding equations.  This simple technique eliminates the need to calculate the OWA weights via the sophisticated O'Hagan’s procedure, which requires the solution of the constrained optimization problem. 35
  • 37. Exponential OWA Operators  Let us assume the number of aggregate objects is 5. Also, assume the desired degree of Orness is 0.6. 36
  • 38. Exponential OWA Operators: An Example  Let us assume the number of aggregate objects is 5. Also, assume the desired degree of Orness is 0.9. 37
  • 39. An Application:Result Fusion in Multi-Agent Systems  In multi-agent systems, each of the agents may have its own expertise.When they are asked to accomplish the same task, the results may be different.  In such situations, it is required to fuse the results to obtain a final one.  This paper [Z. Zhang and C. Zhang, “Result Fusion in Multi-Agent Systems Based on OWA Operator”, 23rd Australasian Computer Science Conference (ACSC 2000), 2000]: – proposes an information fusion approach when multiple agents are asked to collect the information for the same query from different resources. – Discusses about the group decision fusion problem that several agents with similar knowledge make decisions using the same information. 38
  • 40. An Application:Result Fusion in Multi-Agent Systems  Application domain: – Financial Investment Planner Using Intelligent Agent Technologies  Sub-systems: – User interface and result visualization agents – Decision-making agents – Information retrieval and information filter agents – Data mining agents 39
  • 41. Framework of Financial Investment Planner 37