SlideShare a Scribd company logo
1 of 19
CLUSTERING : A NEURAL
NETWORK APPROACH
INDIAN INSTITUTE OF INFORMATION
TECHNOLOGY KALYANI (WB)
ANUJ KUMAR PATHAK
(CSE)
SHIV PRATAP
OUTLINE
• INTRODUCTION
• TYPES OF CLUSTERING
• COMPETITIVE LEANING
• KOHONEN NETWORK(SOM)(COMPETITIVE LEANING BASED
CLUSTERING)
• K-MEAN CLUSTERING OR C -MEAN
• HIERARCHICAL CLUSTERING
• CONCLUSION
INTRODUCTION OF ANN
IT IS A INFORMATION PROCESSING MODEL WORKS LIKE HUMAN
NERVOUS SYSTEM AND CONSIST OF NUMBER OF ARTIFICIAL
NEURONS(PERCEPTRON).
BASIC BUILDING DIAGRAM OF ARTIFICIAL NEURAL NETWORK IS
NEURON WHICH IS MATHEMATICAL MODEL /FUNCTION.
NEURAL NETWORK CLUSTERING
• NEURAL NETWORK CLUSTERING IS A
UNSUPERVISED LEANING.
• BASICALLY CLUSTERING IS THE
ORGANIZATION OF UNLABELLED
DATA E.G. TWEET,PICTURE,VEDIO
INTO SIMILAR GROUP CALLED
CLUSTER.
• CLUSTER IS A COLLECTION OF DATA
ITEM WHICH ARE SIMILAR AND
DISSIMILAR TO DATA ITEM OF
DIFFERENT CLUSTER.
COMPETITIVE LEARNING
• COMPETITIVE LEARNING IS USEFUL FOR CLUSTERING INPUT PATTERNS INTO A
DISCRETE SET OF OUTPUT CLUSTER.
• COMPETITIVE LEARNING CAN BE BUILD USING 2 LAYER (J-K) NEURAL
NETWORK .
• IN COMPETITIVE LEARNING NETWORK INPUT AND OUTPUT ARE FULLY
CONNECTED .
• OUTPUT LAYER IS CALLED COMPETITIVE LAYER.
• WHEREIN LATERAL CONNECTIONS ARE USED TO PERFORM LATERAL
INHIBITION.
Continue..
Competitive is usually derived by minimizing the mean square error(MSE)
E = (1/N)*∑ Ep p=1 to n
where Ep=∑ µkp.|Xp-Ck|.|Xp-Ck|
where n=size of pattern set
and µkp is the connection weight assigned to Ck with respect to Xp
When Ck is the closest prototype Xp then µkp=1 otherwisw 0.
Assumption that weights are obtained by the nearest prototype condition
Thus Ep=min|Xp-Ck|.|Xp-Ck|
Which is the Squared Euclidean distance between the input Xp and its closest prototype Ck.
KOHONEN NETWORK(SOM)
• SOM IS VERY USEFUL FOR VQ CLUSTERING ANALYSIS , FEATURE EXTRACTION AND
DATA VISUALIZATION . IT IS A COMPETITIVE LEARNING BASED MODEL OR ALGORITHM.
• THE GOAL OF SOM IS TO TRANSFORM HIGHER INPUT SPACE INTO 1 1-D OR 2-D
DISCRETE MAP IN A TOPOLOGICAL FASHION.
• SOM CONSIST OF NODES OR NEURONS ASSOCIATED WITH EACH NODE HAVE A
WEIGHT VECTOR OF SAME DIMENSION AS THE INPUT DATA VECTORS AND A POSITION IN
THE MAP SPACE IN HEXAGONAL OR RECTANGULAR AND FORM A NETWORK LIKE
FORWARD NETWORK.
• THIS MAKES SOM USEFUL FOR VISUALIZATION OF LOW DIMENSIONAL VIEW OF HIGH
DIMENSIONAL DATA.
ALGORITHMIC STEPS::
• 1.RANDOMIZE NODE WEIGHT VECTOR IN MAP
• 2.RANDOMLI PICK UP INPUT VECTOR D(T)
• 3.TRAVERSE EACH NODE IN THE MAP
• I . USE EUCLIDEAN DISTANCE FORMULA FOR FINDING
TO FIND SIMILARITY BETWEEN INPUT VECTOR AND
MAP NODE VECTOR.
• II . TRACK THE NODE THAT PRODUCE SMALLEST
DISTANCE(THIS NODE IS KNOWN AS BEST MATCHING
UNIT CALLED BMU)
• III . UPDATE THE NODE IN THE NEIGHBOURHOOD OF
BMU
TYPES OF CLUSTERING
THERE IS THREE TYPES OF CLUSTERING
1 : - PARTITIONAL CLUSTERING (DYNAMIC FORM)
2:- HIERARCHICAL CLUSTERING (STATIC FORM)
3:- DENSITY CLUSTERING
PARTITIONAL CLUSTERING
• ITS SIMPLY A DIVISION OF THE SET OF DATA OBJECTS INTO NON-OVERLAPPING
CLUSTERS SUCH THAT EACH OBJECTS IS IN EXACTLY ONE SUBSET
• IT IS DYNAMIC CLUSTERING
• POINTS CAN MOVE FROM ONE CLUSTERS TO ANOTHER CLUSTER
• KNOWLEDGE OF SHAPE OR SIZE CAN BE INCORPORATED FOR MEASURING
DISTANCE
• IT IS SUSCEPTIBLE TO LOCAL MINIMA OF ITS OBJECTIVE FUNCTION
• NUMBER OF CLUSTER IS TO BE PREDEFINED
• EXAMPLE K-MEAN OR CENTROID CLUSTERING
K MEAN OR CENTROID CLUSTERING
• K MEAN CLUSTERING IS SPECIAL CASE OF SOM
• K MEANS CLUSTERING IS AN UNSUPERVISED LEARNING ALGORITHM THAT
TRIES TO CLUSTER DATA BASED ON THEIR SIMILARITY.
• SPECIFY THE NUMBER OF CLUSTERS WE WANT THE DATA TO BE GROUPED
INTO
• K MEAN CLUSTERING IS TWO TYPES
1:- BATCH MODE :- WHOLE TRAINING DATA SET IS FIXED , NEAREST
NEIGHBOUR RULE
2:- INCREMENTAL MODE :- TRAINING DATA SET CAN BE INCREASE
K MEAN CLUSTERING ALGORITHM
STEP 1:-SPECIFY THE NUMBER OF CLUSTER U WANT
STEP 2:- RANDOMLY ASSIGNS EACH OBSERVATION TO A CLUSTER, AND FINDS THE
CENTROID OF EACH CLUSTER
STEP3:- FOLLOW STEP 4 AND 5 UNTIL NO VARIATION IS FOUND
STEP 4: REASSIGN DATA POINTS TO THE CLUSTER WHOSE CENTROID IS CLOSEST
STEP 5: CALCULATE NEW CENTROID OF EACH CLUSTER (TAKING MEAN OF ALL DATA
IN CLUSTER)
To select the number of clusters we use Elbow method
HIERARCHICAL CLUSTERING
• IT ALSO KNOWN AS 'NESTING
CLUSTERING' AS IT ALSO CLUSTERS
TO EXIST WITHIN BIGGER CLUSTERS
TO FORM A TREE(DENDROME)
• CONSISTS OF A SEQUENCE OF
PARTITIONS IN A HIERARCHICAL
STRUCTURE
• IT IS STATIC (POINT COMMITTED TO
ONE CLUSTERING DOES NOT MOVE
TYPES OF HIERARCHICAL CLUSTERING
• 1:- AGGLOMERATIVE CLUSTERING
MERGING OF ONE CLUSTER TO ANOTHER ON BASIS OF SOME PROPERTY
ALSO KNOW ALSO CALLED BOTTOM UP APPROACH
• 2:- DIVISIVE CLUSTERING
ALL OBSERVATIONS START IN ONE CLUSTER, AND SPLITS ARE PERFORMED
RECURSIVELY AS ONE MOVES DOWN THE HIERARCHY
THIS IS A "TOP DOWN" APPROACH:
AGGLOMERATIVE CLUSTERING
ALGORITHM
STEP 1 START BY ASSIGNING EACH ITEM TO A CLUSTER, SO THAT IF YOU HAVE
N ITEMS (CONSIDER THERE DISTANCE BETWEEN THEM IS SAME)
STEP 2 FIND THE CLOSEST (MOST SIMILAR) PAIR OF CLUSTERS AND MERGE
THEM INTO A SINGLE CLUSTER, SO THAT NOW YOU HAVE ONE CLUSTER
LESS.
STEP 3 COMPUTE DISTANCES (SIMILARITIES) BETWEEN THE NEW CLUSTER AND
EACH OF THE OLD CLUSTERS.
STEP 4 REPEAT STEPS 2 AND 3 UNTIL ALL ITEMS ARE CLUSTERED INTO A
SINGLE CLUSTER OF SIZE N. (*)
THE SELECTION OF 3RD STEP IS RESPONSIBLE FOR DIFFERENT TYPE OF
AGGLOMERATIVE CLUSTERING
CONTINUE…
3RD STEP CAN BE DONE VIA SINGLE LINKAGE , COMPLETE LINKAGE AND
AVERAGE LINKAGE
SINGLE LINKAGE : CALCULATES THE INTER-CLUSTER DISTANCE USING THE
CLOSEST DATA POINTS IN DIFFERENT CLUSTERS
COMPLETE LINKAGE : CALCULATES THE INTER-CLUSTER DISTANCE USING THE
FARTHEST DATA POINTS IN DIFFERENT CLUSTERS
AVERAGE LINKAGE : EQUAL TO THE AVERAGE DISTANCE FROM ANY MEMBER
OF ONE CLUSTER TO ANY MEMBER OF THE OTHER CLUSTER.
COMPARISON BETWEEN K-MEAN
AND HIERARCHICAL CLUSTERING
HIERARCHICAL CLUSTERING
• TIME COMPLEXITY IS O(N*N)
• NO PREDICTION OF NUMBER OF
CLUSTERS
• NO PROBLEM OF LOCAL MINIMUM
INITIALIZATION PROBLEM
• NO PRIOR KNOWLEDGE OF SHAPE
AND SIZE IS REQUIRED
K-MEAN CLUSTERING
• TIME COMPLEXITY IS O(N)
• TIGHTER BOUND
• PREDICT THE VALUE OF NUMBER OF
CLUSTER
• LOCAL MINIMUM INITIALIZATION
PROBLEM
• PRIOR KNOWLEDGE OF SHAPE AND
SIZE IS REQUIRED
Neural Network Clustering Techniques

More Related Content

What's hot

Semantic Segmentation - Fully Convolutional Networks for Semantic Segmentation
Semantic Segmentation - Fully Convolutional Networks for Semantic SegmentationSemantic Segmentation - Fully Convolutional Networks for Semantic Segmentation
Semantic Segmentation - Fully Convolutional Networks for Semantic Segmentation岳華 杜
 
Principal Component Analysis for Tensor Analysis and EEG classification
Principal Component Analysis for Tensor Analysis and EEG classificationPrincipal Component Analysis for Tensor Analysis and EEG classification
Principal Component Analysis for Tensor Analysis and EEG classificationTatsuya Yokota
 
Linear Discriminant Analysis (LDA)
Linear Discriminant Analysis (LDA)Linear Discriminant Analysis (LDA)
Linear Discriminant Analysis (LDA)Anmol Dwivedi
 
Neural Networks: Multilayer Perceptron
Neural Networks: Multilayer PerceptronNeural Networks: Multilayer Perceptron
Neural Networks: Multilayer PerceptronMostafa G. M. Mostafa
 
2.5 backpropagation
2.5 backpropagation2.5 backpropagation
2.5 backpropagationKrish_ver2
 
Introduction to Recurrent Neural Network
Introduction to Recurrent Neural NetworkIntroduction to Recurrent Neural Network
Introduction to Recurrent Neural NetworkKnoldus Inc.
 
Discriminant Analysis-lecture 8
Discriminant Analysis-lecture 8Discriminant Analysis-lecture 8
Discriminant Analysis-lecture 8Laila Fatehy
 
Lecture 4 Decision Trees (2): Entropy, Information Gain, Gain Ratio
Lecture 4 Decision Trees (2): Entropy, Information Gain, Gain RatioLecture 4 Decision Trees (2): Entropy, Information Gain, Gain Ratio
Lecture 4 Decision Trees (2): Entropy, Information Gain, Gain RatioMarina Santini
 
K means clustering
K means clusteringK means clustering
K means clusteringKuppusamy P
 
Neural Networks: Radial Bases Functions (RBF)
Neural Networks: Radial Bases Functions (RBF)Neural Networks: Radial Bases Functions (RBF)
Neural Networks: Radial Bases Functions (RBF)Mostafa G. M. Mostafa
 
Classification Based Machine Learning Algorithms
Classification Based Machine Learning AlgorithmsClassification Based Machine Learning Algorithms
Classification Based Machine Learning AlgorithmsMd. Main Uddin Rony
 
Multilayer perceptron
Multilayer perceptronMultilayer perceptron
Multilayer perceptronomaraldabash
 
Bayseian decision theory
Bayseian decision theoryBayseian decision theory
Bayseian decision theorysia16
 
Principal component analysis and lda
Principal component analysis and ldaPrincipal component analysis and lda
Principal component analysis and ldaSuresh Pokharel
 
Deep Learning - CNN and RNN
Deep Learning - CNN and RNNDeep Learning - CNN and RNN
Deep Learning - CNN and RNNAshray Bhandare
 

What's hot (20)

Semantic Segmentation - Fully Convolutional Networks for Semantic Segmentation
Semantic Segmentation - Fully Convolutional Networks for Semantic SegmentationSemantic Segmentation - Fully Convolutional Networks for Semantic Segmentation
Semantic Segmentation - Fully Convolutional Networks for Semantic Segmentation
 
Principal Component Analysis for Tensor Analysis and EEG classification
Principal Component Analysis for Tensor Analysis and EEG classificationPrincipal Component Analysis for Tensor Analysis and EEG classification
Principal Component Analysis for Tensor Analysis and EEG classification
 
Linear Discriminant Analysis (LDA)
Linear Discriminant Analysis (LDA)Linear Discriminant Analysis (LDA)
Linear Discriminant Analysis (LDA)
 
Neural Networks: Multilayer Perceptron
Neural Networks: Multilayer PerceptronNeural Networks: Multilayer Perceptron
Neural Networks: Multilayer Perceptron
 
Presentation on K-Means Clustering
Presentation on K-Means ClusteringPresentation on K-Means Clustering
Presentation on K-Means Clustering
 
2.5 backpropagation
2.5 backpropagation2.5 backpropagation
2.5 backpropagation
 
Hierachical clustering
Hierachical clusteringHierachical clustering
Hierachical clustering
 
Introduction to Recurrent Neural Network
Introduction to Recurrent Neural NetworkIntroduction to Recurrent Neural Network
Introduction to Recurrent Neural Network
 
Discriminant Analysis-lecture 8
Discriminant Analysis-lecture 8Discriminant Analysis-lecture 8
Discriminant Analysis-lecture 8
 
Lecture 4 Decision Trees (2): Entropy, Information Gain, Gain Ratio
Lecture 4 Decision Trees (2): Entropy, Information Gain, Gain RatioLecture 4 Decision Trees (2): Entropy, Information Gain, Gain Ratio
Lecture 4 Decision Trees (2): Entropy, Information Gain, Gain Ratio
 
Principal Component Analysis
Principal Component AnalysisPrincipal Component Analysis
Principal Component Analysis
 
K means Clustering Algorithm
K means Clustering AlgorithmK means Clustering Algorithm
K means Clustering Algorithm
 
K means clustering
K means clusteringK means clustering
K means clustering
 
Neural Networks: Radial Bases Functions (RBF)
Neural Networks: Radial Bases Functions (RBF)Neural Networks: Radial Bases Functions (RBF)
Neural Networks: Radial Bases Functions (RBF)
 
Density based clustering
Density based clusteringDensity based clustering
Density based clustering
 
Classification Based Machine Learning Algorithms
Classification Based Machine Learning AlgorithmsClassification Based Machine Learning Algorithms
Classification Based Machine Learning Algorithms
 
Multilayer perceptron
Multilayer perceptronMultilayer perceptron
Multilayer perceptron
 
Bayseian decision theory
Bayseian decision theoryBayseian decision theory
Bayseian decision theory
 
Principal component analysis and lda
Principal component analysis and ldaPrincipal component analysis and lda
Principal component analysis and lda
 
Deep Learning - CNN and RNN
Deep Learning - CNN and RNNDeep Learning - CNN and RNN
Deep Learning - CNN and RNN
 

Similar to Neural Network Clustering Techniques

Machine Learning Unit 4 Semester 3 MSc IT Part 2 Mumbai University
Machine Learning Unit 4 Semester 3  MSc IT Part 2 Mumbai UniversityMachine Learning Unit 4 Semester 3  MSc IT Part 2 Mumbai University
Machine Learning Unit 4 Semester 3 MSc IT Part 2 Mumbai UniversityMadhav Mishra
 
K means Clustering Algorithm
K means Clustering AlgorithmK means Clustering Algorithm
K means Clustering AlgorithmSouma Maiti
 
CSA 3702 machine learning module 3
CSA 3702 machine learning module 3CSA 3702 machine learning module 3
CSA 3702 machine learning module 3Nandhini S
 
Advanced database and data mining & clustering concepts
Advanced database and data mining & clustering conceptsAdvanced database and data mining & clustering concepts
Advanced database and data mining & clustering conceptsNithyananthSengottai
 
machine learning - Clustering in R
machine learning - Clustering in Rmachine learning - Clustering in R
machine learning - Clustering in RSudhakar Chavan
 
B.sc biochem i bobi u 3.2 algorithm + blast
B.sc biochem i bobi u 3.2 algorithm + blastB.sc biochem i bobi u 3.2 algorithm + blast
B.sc biochem i bobi u 3.2 algorithm + blastRai University
 
B.sc biochem i bobi u 3.2 algorithm + blast
B.sc biochem i bobi u 3.2 algorithm + blastB.sc biochem i bobi u 3.2 algorithm + blast
B.sc biochem i bobi u 3.2 algorithm + blastRai University
 
DS9 - Clustering.pptx
DS9 - Clustering.pptxDS9 - Clustering.pptx
DS9 - Clustering.pptxJK970901
 
traffic sign detection using deep learning.pptx
traffic sign detection using deep learning.pptxtraffic sign detection using deep learning.pptx
traffic sign detection using deep learning.pptxbrijeshbs2
 
Types of clustering and different types of clustering algorithms
Types of clustering and different types of clustering algorithmsTypes of clustering and different types of clustering algorithms
Types of clustering and different types of clustering algorithmsPrashanth Guntal
 
Network Intrusion Detection System Using Machine Learning and Deep Learning F...
Network Intrusion Detection System Using Machine Learning and Deep Learning F...Network Intrusion Detection System Using Machine Learning and Deep Learning F...
Network Intrusion Detection System Using Machine Learning and Deep Learning F...Leaving A Legacy
 
26-Clustering MTech-2017.ppt
26-Clustering MTech-2017.ppt26-Clustering MTech-2017.ppt
26-Clustering MTech-2017.pptvikassingh569137
 
Performance Analysis of Different Clustering Algorithm
Performance Analysis of Different Clustering AlgorithmPerformance Analysis of Different Clustering Algorithm
Performance Analysis of Different Clustering AlgorithmIOSR Journals
 
ML Module 3 Non Linear Learning.pptx
ML Module 3 Non Linear Learning.pptxML Module 3 Non Linear Learning.pptx
ML Module 3 Non Linear Learning.pptxDebabrataPain1
 
Cluster Analysis
Cluster Analysis Cluster Analysis
Cluster Analysis Baivab Nag
 
CLUSTERING IN DATA MINING.pdf
CLUSTERING IN DATA MINING.pdfCLUSTERING IN DATA MINING.pdf
CLUSTERING IN DATA MINING.pdfSowmyaJyothi3
 

Similar to Neural Network Clustering Techniques (20)

Machine Learning Unit 4 Semester 3 MSc IT Part 2 Mumbai University
Machine Learning Unit 4 Semester 3  MSc IT Part 2 Mumbai UniversityMachine Learning Unit 4 Semester 3  MSc IT Part 2 Mumbai University
Machine Learning Unit 4 Semester 3 MSc IT Part 2 Mumbai University
 
K means Clustering Algorithm
K means Clustering AlgorithmK means Clustering Algorithm
K means Clustering Algorithm
 
CSA 3702 machine learning module 3
CSA 3702 machine learning module 3CSA 3702 machine learning module 3
CSA 3702 machine learning module 3
 
Advanced database and data mining & clustering concepts
Advanced database and data mining & clustering conceptsAdvanced database and data mining & clustering concepts
Advanced database and data mining & clustering concepts
 
Knn
KnnKnn
Knn
 
machine learning - Clustering in R
machine learning - Clustering in Rmachine learning - Clustering in R
machine learning - Clustering in R
 
UNIT_V_Cluster Analysis.pptx
UNIT_V_Cluster Analysis.pptxUNIT_V_Cluster Analysis.pptx
UNIT_V_Cluster Analysis.pptx
 
Deep learning approach for microarray image analysis
Deep learning approach for microarray image analysisDeep learning approach for microarray image analysis
Deep learning approach for microarray image analysis
 
B.sc biochem i bobi u 3.2 algorithm + blast
B.sc biochem i bobi u 3.2 algorithm + blastB.sc biochem i bobi u 3.2 algorithm + blast
B.sc biochem i bobi u 3.2 algorithm + blast
 
B.sc biochem i bobi u 3.2 algorithm + blast
B.sc biochem i bobi u 3.2 algorithm + blastB.sc biochem i bobi u 3.2 algorithm + blast
B.sc biochem i bobi u 3.2 algorithm + blast
 
DS9 - Clustering.pptx
DS9 - Clustering.pptxDS9 - Clustering.pptx
DS9 - Clustering.pptx
 
traffic sign detection using deep learning.pptx
traffic sign detection using deep learning.pptxtraffic sign detection using deep learning.pptx
traffic sign detection using deep learning.pptx
 
Types of clustering and different types of clustering algorithms
Types of clustering and different types of clustering algorithmsTypes of clustering and different types of clustering algorithms
Types of clustering and different types of clustering algorithms
 
Network Intrusion Detection System Using Machine Learning and Deep Learning F...
Network Intrusion Detection System Using Machine Learning and Deep Learning F...Network Intrusion Detection System Using Machine Learning and Deep Learning F...
Network Intrusion Detection System Using Machine Learning and Deep Learning F...
 
26-Clustering MTech-2017.ppt
26-Clustering MTech-2017.ppt26-Clustering MTech-2017.ppt
26-Clustering MTech-2017.ppt
 
Performance Analysis of Different Clustering Algorithm
Performance Analysis of Different Clustering AlgorithmPerformance Analysis of Different Clustering Algorithm
Performance Analysis of Different Clustering Algorithm
 
F017132529
F017132529F017132529
F017132529
 
ML Module 3 Non Linear Learning.pptx
ML Module 3 Non Linear Learning.pptxML Module 3 Non Linear Learning.pptx
ML Module 3 Non Linear Learning.pptx
 
Cluster Analysis
Cluster Analysis Cluster Analysis
Cluster Analysis
 
CLUSTERING IN DATA MINING.pdf
CLUSTERING IN DATA MINING.pdfCLUSTERING IN DATA MINING.pdf
CLUSTERING IN DATA MINING.pdf
 

Recently uploaded

Internship report on mechanical engineering
Internship report on mechanical engineeringInternship report on mechanical engineering
Internship report on mechanical engineeringmalavadedarshan25
 
SPICE PARK APR2024 ( 6,793 SPICE Models )
SPICE PARK APR2024 ( 6,793 SPICE Models )SPICE PARK APR2024 ( 6,793 SPICE Models )
SPICE PARK APR2024 ( 6,793 SPICE Models )Tsuyoshi Horigome
 
OSVC_Meta-Data based Simulation Automation to overcome Verification Challenge...
OSVC_Meta-Data based Simulation Automation to overcome Verification Challenge...OSVC_Meta-Data based Simulation Automation to overcome Verification Challenge...
OSVC_Meta-Data based Simulation Automation to overcome Verification Challenge...Soham Mondal
 
HARMONY IN THE HUMAN BEING - Unit-II UHV-2
HARMONY IN THE HUMAN BEING - Unit-II UHV-2HARMONY IN THE HUMAN BEING - Unit-II UHV-2
HARMONY IN THE HUMAN BEING - Unit-II UHV-2RajaP95
 
complete construction, environmental and economics information of biomass com...
complete construction, environmental and economics information of biomass com...complete construction, environmental and economics information of biomass com...
complete construction, environmental and economics information of biomass com...asadnawaz62
 
Decoding Kotlin - Your guide to solving the mysterious in Kotlin.pptx
Decoding Kotlin - Your guide to solving the mysterious in Kotlin.pptxDecoding Kotlin - Your guide to solving the mysterious in Kotlin.pptx
Decoding Kotlin - Your guide to solving the mysterious in Kotlin.pptxJoão Esperancinha
 
power system scada applications and uses
power system scada applications and usespower system scada applications and uses
power system scada applications and usesDevarapalliHaritha
 
Call Girls Delhi {Jodhpur} 9711199012 high profile service
Call Girls Delhi {Jodhpur} 9711199012 high profile serviceCall Girls Delhi {Jodhpur} 9711199012 high profile service
Call Girls Delhi {Jodhpur} 9711199012 high profile servicerehmti665
 
IVE Industry Focused Event - Defence Sector 2024
IVE Industry Focused Event - Defence Sector 2024IVE Industry Focused Event - Defence Sector 2024
IVE Industry Focused Event - Defence Sector 2024Mark Billinghurst
 
Artificial-Intelligence-in-Electronics (K).pptx
Artificial-Intelligence-in-Electronics (K).pptxArtificial-Intelligence-in-Electronics (K).pptx
Artificial-Intelligence-in-Electronics (K).pptxbritheesh05
 
main PPT.pptx of girls hostel security using rfid
main PPT.pptx of girls hostel security using rfidmain PPT.pptx of girls hostel security using rfid
main PPT.pptx of girls hostel security using rfidNikhilNagaraju
 
What are the advantages and disadvantages of membrane structures.pptx
What are the advantages and disadvantages of membrane structures.pptxWhat are the advantages and disadvantages of membrane structures.pptx
What are the advantages and disadvantages of membrane structures.pptxwendy cai
 
INFLUENCE OF NANOSILICA ON THE PROPERTIES OF CONCRETE
INFLUENCE OF NANOSILICA ON THE PROPERTIES OF CONCRETEINFLUENCE OF NANOSILICA ON THE PROPERTIES OF CONCRETE
INFLUENCE OF NANOSILICA ON THE PROPERTIES OF CONCRETEroselinkalist12
 
CCS355 Neural Network & Deep Learning UNIT III notes and Question bank .pdf
CCS355 Neural Network & Deep Learning UNIT III notes and Question bank .pdfCCS355 Neural Network & Deep Learning UNIT III notes and Question bank .pdf
CCS355 Neural Network & Deep Learning UNIT III notes and Question bank .pdfAsst.prof M.Gokilavani
 
Gurgaon ✡️9711147426✨Call In girls Gurgaon Sector 51 escort service
Gurgaon ✡️9711147426✨Call In girls Gurgaon Sector 51 escort serviceGurgaon ✡️9711147426✨Call In girls Gurgaon Sector 51 escort service
Gurgaon ✡️9711147426✨Call In girls Gurgaon Sector 51 escort servicejennyeacort
 
Study on Air-Water & Water-Water Heat Exchange in a Finned Tube Exchanger
Study on Air-Water & Water-Water Heat Exchange in a Finned Tube ExchangerStudy on Air-Water & Water-Water Heat Exchange in a Finned Tube Exchanger
Study on Air-Water & Water-Water Heat Exchange in a Finned Tube ExchangerAnamika Sarkar
 
chaitra-1.pptx fake news detection using machine learning
chaitra-1.pptx  fake news detection using machine learningchaitra-1.pptx  fake news detection using machine learning
chaitra-1.pptx fake news detection using machine learningmisbanausheenparvam
 
Architect Hassan Khalil Portfolio for 2024
Architect Hassan Khalil Portfolio for 2024Architect Hassan Khalil Portfolio for 2024
Architect Hassan Khalil Portfolio for 2024hassan khalil
 

Recently uploaded (20)

Internship report on mechanical engineering
Internship report on mechanical engineeringInternship report on mechanical engineering
Internship report on mechanical engineering
 
SPICE PARK APR2024 ( 6,793 SPICE Models )
SPICE PARK APR2024 ( 6,793 SPICE Models )SPICE PARK APR2024 ( 6,793 SPICE Models )
SPICE PARK APR2024 ( 6,793 SPICE Models )
 
OSVC_Meta-Data based Simulation Automation to overcome Verification Challenge...
OSVC_Meta-Data based Simulation Automation to overcome Verification Challenge...OSVC_Meta-Data based Simulation Automation to overcome Verification Challenge...
OSVC_Meta-Data based Simulation Automation to overcome Verification Challenge...
 
HARMONY IN THE HUMAN BEING - Unit-II UHV-2
HARMONY IN THE HUMAN BEING - Unit-II UHV-2HARMONY IN THE HUMAN BEING - Unit-II UHV-2
HARMONY IN THE HUMAN BEING - Unit-II UHV-2
 
complete construction, environmental and economics information of biomass com...
complete construction, environmental and economics information of biomass com...complete construction, environmental and economics information of biomass com...
complete construction, environmental and economics information of biomass com...
 
Decoding Kotlin - Your guide to solving the mysterious in Kotlin.pptx
Decoding Kotlin - Your guide to solving the mysterious in Kotlin.pptxDecoding Kotlin - Your guide to solving the mysterious in Kotlin.pptx
Decoding Kotlin - Your guide to solving the mysterious in Kotlin.pptx
 
power system scada applications and uses
power system scada applications and usespower system scada applications and uses
power system scada applications and uses
 
Call Girls Delhi {Jodhpur} 9711199012 high profile service
Call Girls Delhi {Jodhpur} 9711199012 high profile serviceCall Girls Delhi {Jodhpur} 9711199012 high profile service
Call Girls Delhi {Jodhpur} 9711199012 high profile service
 
IVE Industry Focused Event - Defence Sector 2024
IVE Industry Focused Event - Defence Sector 2024IVE Industry Focused Event - Defence Sector 2024
IVE Industry Focused Event - Defence Sector 2024
 
Artificial-Intelligence-in-Electronics (K).pptx
Artificial-Intelligence-in-Electronics (K).pptxArtificial-Intelligence-in-Electronics (K).pptx
Artificial-Intelligence-in-Electronics (K).pptx
 
main PPT.pptx of girls hostel security using rfid
main PPT.pptx of girls hostel security using rfidmain PPT.pptx of girls hostel security using rfid
main PPT.pptx of girls hostel security using rfid
 
Call Us -/9953056974- Call Girls In Vikaspuri-/- Delhi NCR
Call Us -/9953056974- Call Girls In Vikaspuri-/- Delhi NCRCall Us -/9953056974- Call Girls In Vikaspuri-/- Delhi NCR
Call Us -/9953056974- Call Girls In Vikaspuri-/- Delhi NCR
 
What are the advantages and disadvantages of membrane structures.pptx
What are the advantages and disadvantages of membrane structures.pptxWhat are the advantages and disadvantages of membrane structures.pptx
What are the advantages and disadvantages of membrane structures.pptx
 
INFLUENCE OF NANOSILICA ON THE PROPERTIES OF CONCRETE
INFLUENCE OF NANOSILICA ON THE PROPERTIES OF CONCRETEINFLUENCE OF NANOSILICA ON THE PROPERTIES OF CONCRETE
INFLUENCE OF NANOSILICA ON THE PROPERTIES OF CONCRETE
 
CCS355 Neural Network & Deep Learning UNIT III notes and Question bank .pdf
CCS355 Neural Network & Deep Learning UNIT III notes and Question bank .pdfCCS355 Neural Network & Deep Learning UNIT III notes and Question bank .pdf
CCS355 Neural Network & Deep Learning UNIT III notes and Question bank .pdf
 
★ CALL US 9953330565 ( HOT Young Call Girls In Badarpur delhi NCR
★ CALL US 9953330565 ( HOT Young Call Girls In Badarpur delhi NCR★ CALL US 9953330565 ( HOT Young Call Girls In Badarpur delhi NCR
★ CALL US 9953330565 ( HOT Young Call Girls In Badarpur delhi NCR
 
Gurgaon ✡️9711147426✨Call In girls Gurgaon Sector 51 escort service
Gurgaon ✡️9711147426✨Call In girls Gurgaon Sector 51 escort serviceGurgaon ✡️9711147426✨Call In girls Gurgaon Sector 51 escort service
Gurgaon ✡️9711147426✨Call In girls Gurgaon Sector 51 escort service
 
Study on Air-Water & Water-Water Heat Exchange in a Finned Tube Exchanger
Study on Air-Water & Water-Water Heat Exchange in a Finned Tube ExchangerStudy on Air-Water & Water-Water Heat Exchange in a Finned Tube Exchanger
Study on Air-Water & Water-Water Heat Exchange in a Finned Tube Exchanger
 
chaitra-1.pptx fake news detection using machine learning
chaitra-1.pptx  fake news detection using machine learningchaitra-1.pptx  fake news detection using machine learning
chaitra-1.pptx fake news detection using machine learning
 
Architect Hassan Khalil Portfolio for 2024
Architect Hassan Khalil Portfolio for 2024Architect Hassan Khalil Portfolio for 2024
Architect Hassan Khalil Portfolio for 2024
 

Neural Network Clustering Techniques

  • 1. CLUSTERING : A NEURAL NETWORK APPROACH INDIAN INSTITUTE OF INFORMATION TECHNOLOGY KALYANI (WB) ANUJ KUMAR PATHAK (CSE) SHIV PRATAP
  • 2. OUTLINE • INTRODUCTION • TYPES OF CLUSTERING • COMPETITIVE LEANING • KOHONEN NETWORK(SOM)(COMPETITIVE LEANING BASED CLUSTERING) • K-MEAN CLUSTERING OR C -MEAN • HIERARCHICAL CLUSTERING • CONCLUSION
  • 3. INTRODUCTION OF ANN IT IS A INFORMATION PROCESSING MODEL WORKS LIKE HUMAN NERVOUS SYSTEM AND CONSIST OF NUMBER OF ARTIFICIAL NEURONS(PERCEPTRON). BASIC BUILDING DIAGRAM OF ARTIFICIAL NEURAL NETWORK IS NEURON WHICH IS MATHEMATICAL MODEL /FUNCTION.
  • 4. NEURAL NETWORK CLUSTERING • NEURAL NETWORK CLUSTERING IS A UNSUPERVISED LEANING. • BASICALLY CLUSTERING IS THE ORGANIZATION OF UNLABELLED DATA E.G. TWEET,PICTURE,VEDIO INTO SIMILAR GROUP CALLED CLUSTER. • CLUSTER IS A COLLECTION OF DATA ITEM WHICH ARE SIMILAR AND DISSIMILAR TO DATA ITEM OF DIFFERENT CLUSTER.
  • 5. COMPETITIVE LEARNING • COMPETITIVE LEARNING IS USEFUL FOR CLUSTERING INPUT PATTERNS INTO A DISCRETE SET OF OUTPUT CLUSTER. • COMPETITIVE LEARNING CAN BE BUILD USING 2 LAYER (J-K) NEURAL NETWORK . • IN COMPETITIVE LEARNING NETWORK INPUT AND OUTPUT ARE FULLY CONNECTED . • OUTPUT LAYER IS CALLED COMPETITIVE LAYER. • WHEREIN LATERAL CONNECTIONS ARE USED TO PERFORM LATERAL INHIBITION.
  • 6. Continue.. Competitive is usually derived by minimizing the mean square error(MSE) E = (1/N)*∑ Ep p=1 to n where Ep=∑ µkp.|Xp-Ck|.|Xp-Ck| where n=size of pattern set and µkp is the connection weight assigned to Ck with respect to Xp When Ck is the closest prototype Xp then µkp=1 otherwisw 0. Assumption that weights are obtained by the nearest prototype condition Thus Ep=min|Xp-Ck|.|Xp-Ck| Which is the Squared Euclidean distance between the input Xp and its closest prototype Ck.
  • 7. KOHONEN NETWORK(SOM) • SOM IS VERY USEFUL FOR VQ CLUSTERING ANALYSIS , FEATURE EXTRACTION AND DATA VISUALIZATION . IT IS A COMPETITIVE LEARNING BASED MODEL OR ALGORITHM. • THE GOAL OF SOM IS TO TRANSFORM HIGHER INPUT SPACE INTO 1 1-D OR 2-D DISCRETE MAP IN A TOPOLOGICAL FASHION. • SOM CONSIST OF NODES OR NEURONS ASSOCIATED WITH EACH NODE HAVE A WEIGHT VECTOR OF SAME DIMENSION AS THE INPUT DATA VECTORS AND A POSITION IN THE MAP SPACE IN HEXAGONAL OR RECTANGULAR AND FORM A NETWORK LIKE FORWARD NETWORK. • THIS MAKES SOM USEFUL FOR VISUALIZATION OF LOW DIMENSIONAL VIEW OF HIGH DIMENSIONAL DATA.
  • 8. ALGORITHMIC STEPS:: • 1.RANDOMIZE NODE WEIGHT VECTOR IN MAP • 2.RANDOMLI PICK UP INPUT VECTOR D(T) • 3.TRAVERSE EACH NODE IN THE MAP • I . USE EUCLIDEAN DISTANCE FORMULA FOR FINDING TO FIND SIMILARITY BETWEEN INPUT VECTOR AND MAP NODE VECTOR. • II . TRACK THE NODE THAT PRODUCE SMALLEST DISTANCE(THIS NODE IS KNOWN AS BEST MATCHING UNIT CALLED BMU) • III . UPDATE THE NODE IN THE NEIGHBOURHOOD OF BMU
  • 9. TYPES OF CLUSTERING THERE IS THREE TYPES OF CLUSTERING 1 : - PARTITIONAL CLUSTERING (DYNAMIC FORM) 2:- HIERARCHICAL CLUSTERING (STATIC FORM) 3:- DENSITY CLUSTERING
  • 10. PARTITIONAL CLUSTERING • ITS SIMPLY A DIVISION OF THE SET OF DATA OBJECTS INTO NON-OVERLAPPING CLUSTERS SUCH THAT EACH OBJECTS IS IN EXACTLY ONE SUBSET • IT IS DYNAMIC CLUSTERING • POINTS CAN MOVE FROM ONE CLUSTERS TO ANOTHER CLUSTER • KNOWLEDGE OF SHAPE OR SIZE CAN BE INCORPORATED FOR MEASURING DISTANCE • IT IS SUSCEPTIBLE TO LOCAL MINIMA OF ITS OBJECTIVE FUNCTION • NUMBER OF CLUSTER IS TO BE PREDEFINED • EXAMPLE K-MEAN OR CENTROID CLUSTERING
  • 11. K MEAN OR CENTROID CLUSTERING • K MEAN CLUSTERING IS SPECIAL CASE OF SOM • K MEANS CLUSTERING IS AN UNSUPERVISED LEARNING ALGORITHM THAT TRIES TO CLUSTER DATA BASED ON THEIR SIMILARITY. • SPECIFY THE NUMBER OF CLUSTERS WE WANT THE DATA TO BE GROUPED INTO • K MEAN CLUSTERING IS TWO TYPES 1:- BATCH MODE :- WHOLE TRAINING DATA SET IS FIXED , NEAREST NEIGHBOUR RULE 2:- INCREMENTAL MODE :- TRAINING DATA SET CAN BE INCREASE
  • 12. K MEAN CLUSTERING ALGORITHM STEP 1:-SPECIFY THE NUMBER OF CLUSTER U WANT STEP 2:- RANDOMLY ASSIGNS EACH OBSERVATION TO A CLUSTER, AND FINDS THE CENTROID OF EACH CLUSTER STEP3:- FOLLOW STEP 4 AND 5 UNTIL NO VARIATION IS FOUND STEP 4: REASSIGN DATA POINTS TO THE CLUSTER WHOSE CENTROID IS CLOSEST STEP 5: CALCULATE NEW CENTROID OF EACH CLUSTER (TAKING MEAN OF ALL DATA IN CLUSTER)
  • 13. To select the number of clusters we use Elbow method
  • 14. HIERARCHICAL CLUSTERING • IT ALSO KNOWN AS 'NESTING CLUSTERING' AS IT ALSO CLUSTERS TO EXIST WITHIN BIGGER CLUSTERS TO FORM A TREE(DENDROME) • CONSISTS OF A SEQUENCE OF PARTITIONS IN A HIERARCHICAL STRUCTURE • IT IS STATIC (POINT COMMITTED TO ONE CLUSTERING DOES NOT MOVE
  • 15. TYPES OF HIERARCHICAL CLUSTERING • 1:- AGGLOMERATIVE CLUSTERING MERGING OF ONE CLUSTER TO ANOTHER ON BASIS OF SOME PROPERTY ALSO KNOW ALSO CALLED BOTTOM UP APPROACH • 2:- DIVISIVE CLUSTERING ALL OBSERVATIONS START IN ONE CLUSTER, AND SPLITS ARE PERFORMED RECURSIVELY AS ONE MOVES DOWN THE HIERARCHY THIS IS A "TOP DOWN" APPROACH:
  • 16. AGGLOMERATIVE CLUSTERING ALGORITHM STEP 1 START BY ASSIGNING EACH ITEM TO A CLUSTER, SO THAT IF YOU HAVE N ITEMS (CONSIDER THERE DISTANCE BETWEEN THEM IS SAME) STEP 2 FIND THE CLOSEST (MOST SIMILAR) PAIR OF CLUSTERS AND MERGE THEM INTO A SINGLE CLUSTER, SO THAT NOW YOU HAVE ONE CLUSTER LESS. STEP 3 COMPUTE DISTANCES (SIMILARITIES) BETWEEN THE NEW CLUSTER AND EACH OF THE OLD CLUSTERS. STEP 4 REPEAT STEPS 2 AND 3 UNTIL ALL ITEMS ARE CLUSTERED INTO A SINGLE CLUSTER OF SIZE N. (*) THE SELECTION OF 3RD STEP IS RESPONSIBLE FOR DIFFERENT TYPE OF AGGLOMERATIVE CLUSTERING
  • 17. CONTINUE… 3RD STEP CAN BE DONE VIA SINGLE LINKAGE , COMPLETE LINKAGE AND AVERAGE LINKAGE SINGLE LINKAGE : CALCULATES THE INTER-CLUSTER DISTANCE USING THE CLOSEST DATA POINTS IN DIFFERENT CLUSTERS COMPLETE LINKAGE : CALCULATES THE INTER-CLUSTER DISTANCE USING THE FARTHEST DATA POINTS IN DIFFERENT CLUSTERS AVERAGE LINKAGE : EQUAL TO THE AVERAGE DISTANCE FROM ANY MEMBER OF ONE CLUSTER TO ANY MEMBER OF THE OTHER CLUSTER.
  • 18. COMPARISON BETWEEN K-MEAN AND HIERARCHICAL CLUSTERING HIERARCHICAL CLUSTERING • TIME COMPLEXITY IS O(N*N) • NO PREDICTION OF NUMBER OF CLUSTERS • NO PROBLEM OF LOCAL MINIMUM INITIALIZATION PROBLEM • NO PRIOR KNOWLEDGE OF SHAPE AND SIZE IS REQUIRED K-MEAN CLUSTERING • TIME COMPLEXITY IS O(N) • TIGHTER BOUND • PREDICT THE VALUE OF NUMBER OF CLUSTER • LOCAL MINIMUM INITIALIZATION PROBLEM • PRIOR KNOWLEDGE OF SHAPE AND SIZE IS REQUIRED