SlideShare a Scribd company logo
1 of 12
Download to read offline
Multi-Label Classification with
Meta Labels
Jesse Read(1), Antti Puurula(2), Albert Bifet(3)
(1) Aalto University and HIIT, Finland
(2) University of Waikato, New Zealand
(3) Huawei Noah’s Ark Lab, Hong Kong
17 December 2014
Overview
Multi-label classification:
⊆ {beach, sunset, foliage, field, mountain, urban}
• Most multi-label classification methods can be expressed
in a general framework of meta-label classification
• Our work combines labels into meta-labels, so as to learn
dependence efficiently and effectively.
Multi-label Learning
With input variables X, produce predictions for multiple
output variables Y . The basic binary relevance approach,
YCYBYA
X5X4X3X2X1
• does not capture label dependence (among Y -variables)
• does not scale to large number of labels
The label powerset approach models label combinations as
class values in a multi-class problem.
Meta Labels
We introduce a layer of meta-labels,
YCYBYA
YB,CYA,B
X5X4X3X2X1
• captures label dependence
• meta-labels can be fewer than the original number of
labels
• and are deterministically decodable into the labels
Producing Meta Labels
dataset
instance labels
1 B
2 B,C
3 C
4 B
5 A,C
6 A,C
7 A,C
8 A,B,C
9 C
binary relevance
YA YB YC
0 1 0
0 1 1
0 0 1
0 1 0
1 0 1
1 0 1
1 0 1
1 1 1
0 0 1
label powerset
YA,B,C
B
BC
C
B
AC
AC
AC
ABC
C
meta labels
YA,C YB,C
∅ B
∅ BC
∅ C
∅ B
AC C
AC C
AC C
∅ BC
∅ C
• binary relevance: 9 exs, 3 × 2 binary classes
• label powerset: 9 exs, 1 × 5 multi-class
Producing Meta Labels
dataset
instance labels
1 B
2 B,C
3 C
4 B
5 A,C
6 A,C
7 A,C
8 A,B,C
9 C
binary relevance
YA YB YC
0 1 0
0 1 1
0 0 1
0 1 0
1 0 1
1 0 1
1 0 1
1 1 1
0 0 1
label powerset
YA,B,C
B
BC
C
B
AC
AC
AC
ABC
C
meta labels
YA,C YB,C
∅ B
∅ BC
∅ C
∅ B
AC C
AC C
AC C
∅ BC
∅ C
• binary relevance: 9 exs, 3 × 2 binary classes
• label powerset: 9 exs, 1 × 5 multi-class
• pruned meta labels: 9 exs, 2 meta labels, of 2 and 3 values
YAC ∈ {∅, AC}, YBC ∈ {B, C, BC}
(one possible formulation)
Example 2
There is no need to model labels together if there is no strong
dependence between them,
YCYBYA
YCYA,B
X1X2X3X4X5
YA,B ∈ {∅, B, AB}, YC ∈ {∅, C}
(e.g., no strong relation between C and the other labels)
General process for
classification with meta-labels
1 Make a partition (either overlapping or disjoint) of the
label set
2 Relabel the meta-labels, deciding on how many values
each label can take (i.e., possibly pruning some)
3 Train classifiers to predict meta-labels from the input
instances
4 Make predictions into the meta-label space
5 Recombine predictions into the label space
Voting Using Meta Labels
Table : Meta-label Vote, e.g., for YAB ∈ {∅, B, AB}
v A B p(YAB = v|˜x)
∅ 0 0 0.0
B 0 1 0.9
AB 1 1 0.1
PAB 0.1 1.0
Table : Labelset Voting: From Meta-labels to Labels
A B C
PAB 0.1 1.0
PBC 0.7 0.3
k Pk,j 0.1 1.7 0.3
ˆy (> 0.5) 0 1 0
Results
0 5 10 15 20 25 30 35 400.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
Accuracy
RAkEL
RAkELp
0 5 10 15 20 25 30 35 400
500
1000
1500
2000
Runningtime
RAkEL
RAkELp
Figure : On Enron (1700 emails, 53 labels). RAkEL: random
ensembles of ‘label-powerset method’ on subsets of size k
(horizontal axis) vs RAkELp: with pruned meta-labels
Summary
• General framework of meta-labels for multi-label
classification
• Unifies various approaches from the literature
• New models RAkELp and EpRd have large improvement
in running time
• Part of solution for LSHTC4 (1st place) and WISE (2nd
Place) Kaggle challenges
• Code available at
http://meka.sourceforge.net
Multi-Label Classification with
Meta Labels
Jesse Read(1), Antti Puurula(2), Albert Bifet(3)
(1) Aalto University and HIIT, Finland
(2) University of Waikato, New Zealand
(3) Huawei Noah’s Ark Lab, Hong Kong
17 December 2014

More Related Content

What's hot

Data preprocessing in Data Mining
Data preprocessing in Data MiningData preprocessing in Data Mining
Data preprocessing in Data MiningDHIVYADEVAKI
 
Data Mining: Concepts and Techniques chapter 07 : Advanced Frequent Pattern M...
Data Mining: Concepts and Techniques chapter 07 : Advanced Frequent Pattern M...Data Mining: Concepts and Techniques chapter 07 : Advanced Frequent Pattern M...
Data Mining: Concepts and Techniques chapter 07 : Advanced Frequent Pattern M...Salah Amean
 
Data Mining: Concepts and Techniques (3rd ed.) — Chapter _04 olap
Data Mining:  Concepts and Techniques (3rd ed.)— Chapter _04 olapData Mining:  Concepts and Techniques (3rd ed.)— Chapter _04 olap
Data Mining: Concepts and Techniques (3rd ed.) — Chapter _04 olapSalah Amean
 
Decision trees in Machine Learning
Decision trees in Machine Learning Decision trees in Machine Learning
Decision trees in Machine Learning Mohammad Junaid Khan
 
Data Visualisation.pdf
Data Visualisation.pdfData Visualisation.pdf
Data Visualisation.pdfThiyagu K
 
Steganography and watermarking
Steganography and watermarkingSteganography and watermarking
Steganography and watermarkingsudip nandi
 
Apache Spark - Dataframes & Spark SQL - Part 1 | Big Data Hadoop Spark Tutori...
Apache Spark - Dataframes & Spark SQL - Part 1 | Big Data Hadoop Spark Tutori...Apache Spark - Dataframes & Spark SQL - Part 1 | Big Data Hadoop Spark Tutori...
Apache Spark - Dataframes & Spark SQL - Part 1 | Big Data Hadoop Spark Tutori...CloudxLab
 
DATA WAREHOUSE IMPLEMENTATION BY SAIKIRAN PANJALA
DATA WAREHOUSE IMPLEMENTATION BY SAIKIRAN PANJALADATA WAREHOUSE IMPLEMENTATION BY SAIKIRAN PANJALA
DATA WAREHOUSE IMPLEMENTATION BY SAIKIRAN PANJALASaikiran Panjala
 
Exploratory data analysis in R - Data Science Club
Exploratory data analysis in R - Data Science ClubExploratory data analysis in R - Data Science Club
Exploratory data analysis in R - Data Science ClubMartin Bago
 
Data Mining: clustering and analysis
Data Mining: clustering and analysisData Mining: clustering and analysis
Data Mining: clustering and analysisDataminingTools Inc
 
Data scientist roadmap
Data scientist roadmapData scientist roadmap
Data scientist roadmapSonu Kumar
 
Data Mining:Concepts and Techniques, Chapter 8. Classification: Basic Concepts
Data Mining:Concepts and Techniques, Chapter 8. Classification: Basic ConceptsData Mining:Concepts and Techniques, Chapter 8. Classification: Basic Concepts
Data Mining:Concepts and Techniques, Chapter 8. Classification: Basic ConceptsSalah Amean
 
Chapter 9. Classification Advanced Methods.ppt
Chapter 9. Classification Advanced Methods.pptChapter 9. Classification Advanced Methods.ppt
Chapter 9. Classification Advanced Methods.pptSubrata Kumer Paul
 
Big data visualization
Big data visualizationBig data visualization
Big data visualizationAnurag Gupta
 

What's hot (20)

Big Data
Big DataBig Data
Big Data
 
Data preprocessing in Data Mining
Data preprocessing in Data MiningData preprocessing in Data Mining
Data preprocessing in Data Mining
 
Data Mining: Concepts and Techniques chapter 07 : Advanced Frequent Pattern M...
Data Mining: Concepts and Techniques chapter 07 : Advanced Frequent Pattern M...Data Mining: Concepts and Techniques chapter 07 : Advanced Frequent Pattern M...
Data Mining: Concepts and Techniques chapter 07 : Advanced Frequent Pattern M...
 
Data Mining: Concepts and Techniques (3rd ed.) — Chapter _04 olap
Data Mining:  Concepts and Techniques (3rd ed.)— Chapter _04 olapData Mining:  Concepts and Techniques (3rd ed.)— Chapter _04 olap
Data Mining: Concepts and Techniques (3rd ed.) — Chapter _04 olap
 
Decision trees in Machine Learning
Decision trees in Machine Learning Decision trees in Machine Learning
Decision trees in Machine Learning
 
Data Visualisation.pdf
Data Visualisation.pdfData Visualisation.pdf
Data Visualisation.pdf
 
Big table
Big tableBig table
Big table
 
Steganography and watermarking
Steganography and watermarkingSteganography and watermarking
Steganography and watermarking
 
Apache Spark - Dataframes & Spark SQL - Part 1 | Big Data Hadoop Spark Tutori...
Apache Spark - Dataframes & Spark SQL - Part 1 | Big Data Hadoop Spark Tutori...Apache Spark - Dataframes & Spark SQL - Part 1 | Big Data Hadoop Spark Tutori...
Apache Spark - Dataframes & Spark SQL - Part 1 | Big Data Hadoop Spark Tutori...
 
Kdd process
Kdd processKdd process
Kdd process
 
Ch3 physical layer.ppt
Ch3 physical layer.pptCh3 physical layer.ppt
Ch3 physical layer.ppt
 
DATA WAREHOUSE IMPLEMENTATION BY SAIKIRAN PANJALA
DATA WAREHOUSE IMPLEMENTATION BY SAIKIRAN PANJALADATA WAREHOUSE IMPLEMENTATION BY SAIKIRAN PANJALA
DATA WAREHOUSE IMPLEMENTATION BY SAIKIRAN PANJALA
 
Exploratory data analysis in R - Data Science Club
Exploratory data analysis in R - Data Science ClubExploratory data analysis in R - Data Science Club
Exploratory data analysis in R - Data Science Club
 
Data Mining: clustering and analysis
Data Mining: clustering and analysisData Mining: clustering and analysis
Data Mining: clustering and analysis
 
Data scientist roadmap
Data scientist roadmapData scientist roadmap
Data scientist roadmap
 
Data Mining:Concepts and Techniques, Chapter 8. Classification: Basic Concepts
Data Mining:Concepts and Techniques, Chapter 8. Classification: Basic ConceptsData Mining:Concepts and Techniques, Chapter 8. Classification: Basic Concepts
Data Mining:Concepts and Techniques, Chapter 8. Classification: Basic Concepts
 
Chapter 9. Classification Advanced Methods.ppt
Chapter 9. Classification Advanced Methods.pptChapter 9. Classification Advanced Methods.ppt
Chapter 9. Classification Advanced Methods.ppt
 
Big data visualization
Big data visualizationBig data visualization
Big data visualization
 
Data reduction
Data reductionData reduction
Data reduction
 
Encodings
EncodingsEncodings
Encodings
 

Viewers also liked

MrKNN_Soft Relevance for Multi-label Classification
MrKNN_Soft Relevance for Multi-label ClassificationMrKNN_Soft Relevance for Multi-label Classification
MrKNN_Soft Relevance for Multi-label ClassificationYI-JHEN LIN
 
Tag Extraction Final Presentation - CS185CSpring2014
Tag Extraction Final Presentation - CS185CSpring2014Tag Extraction Final Presentation - CS185CSpring2014
Tag Extraction Final Presentation - CS185CSpring2014Naoki Nakatani
 
Multi-label, Multi-class Classification Using Polylingual Embeddings
Multi-label, Multi-class Classification Using Polylingual EmbeddingsMulti-label, Multi-class Classification Using Polylingual Embeddings
Multi-label, Multi-class Classification Using Polylingual EmbeddingsGeorge Balikas
 
Multi-Class Classification on Cartographic Data(Forest Cover)
Multi-Class Classification on Cartographic Data(Forest Cover)Multi-Class Classification on Cartographic Data(Forest Cover)
Multi-Class Classification on Cartographic Data(Forest Cover)Abhishek Agrawal
 
Voting Based Learning Classifier System for Multi-Label Classification
Voting Based Learning Classifier System for Multi-Label ClassificationVoting Based Learning Classifier System for Multi-Label Classification
Voting Based Learning Classifier System for Multi-Label ClassificationDaniele Loiacono
 
Svm implementation for Health Data
Svm implementation for Health DataSvm implementation for Health Data
Svm implementation for Health DataAbhishek Agrawal
 
CNN-RNN: A Unified Framework for Multi-label Image Classification@CV勉強会35回CVP...
CNN-RNN: A Unified Framework for Multi-label Image Classification@CV勉強会35回CVP...CNN-RNN: A Unified Framework for Multi-label Image Classification@CV勉強会35回CVP...
CNN-RNN: A Unified Framework for Multi-label Image Classification@CV勉強会35回CVP...Toshiki Sakai
 
Naïve multi label classification of you tube comments using
Naïve multi label classification of you tube comments usingNaïve multi label classification of you tube comments using
Naïve multi label classification of you tube comments usingNidhi Baranwal
 
Efficient Online Evaluation of Big Data Stream Classifiers
Efficient Online Evaluation of Big Data Stream ClassifiersEfficient Online Evaluation of Big Data Stream Classifiers
Efficient Online Evaluation of Big Data Stream ClassifiersAlbert Bifet
 
Data Classification Presentation
Data Classification PresentationData Classification Presentation
Data Classification PresentationDerroylo
 
Introduction to Text Mining
Introduction to Text MiningIntroduction to Text Mining
Introduction to Text MiningMinha Hwang
 
Introduction to Sentiment Analysis
Introduction to Sentiment AnalysisIntroduction to Sentiment Analysis
Introduction to Sentiment AnalysisJaganadh Gopinadhan
 

Viewers also liked (15)

MrKNN_Soft Relevance for Multi-label Classification
MrKNN_Soft Relevance for Multi-label ClassificationMrKNN_Soft Relevance for Multi-label Classification
MrKNN_Soft Relevance for Multi-label Classification
 
Tag Extraction Final Presentation - CS185CSpring2014
Tag Extraction Final Presentation - CS185CSpring2014Tag Extraction Final Presentation - CS185CSpring2014
Tag Extraction Final Presentation - CS185CSpring2014
 
Multi-label, Multi-class Classification Using Polylingual Embeddings
Multi-label, Multi-class Classification Using Polylingual EmbeddingsMulti-label, Multi-class Classification Using Polylingual Embeddings
Multi-label, Multi-class Classification Using Polylingual Embeddings
 
Multi-Class Classification on Cartographic Data(Forest Cover)
Multi-Class Classification on Cartographic Data(Forest Cover)Multi-Class Classification on Cartographic Data(Forest Cover)
Multi-Class Classification on Cartographic Data(Forest Cover)
 
Voting Based Learning Classifier System for Multi-Label Classification
Voting Based Learning Classifier System for Multi-Label ClassificationVoting Based Learning Classifier System for Multi-Label Classification
Voting Based Learning Classifier System for Multi-Label Classification
 
Svm implementation for Health Data
Svm implementation for Health DataSvm implementation for Health Data
Svm implementation for Health Data
 
CNN-RNN: A Unified Framework for Multi-label Image Classification@CV勉強会35回CVP...
CNN-RNN: A Unified Framework for Multi-label Image Classification@CV勉強会35回CVP...CNN-RNN: A Unified Framework for Multi-label Image Classification@CV勉強会35回CVP...
CNN-RNN: A Unified Framework for Multi-label Image Classification@CV勉強会35回CVP...
 
Naïve multi label classification of you tube comments using
Naïve multi label classification of you tube comments usingNaïve multi label classification of you tube comments using
Naïve multi label classification of you tube comments using
 
Efficient Online Evaluation of Big Data Stream Classifiers
Efficient Online Evaluation of Big Data Stream ClassifiersEfficient Online Evaluation of Big Data Stream Classifiers
Efficient Online Evaluation of Big Data Stream Classifiers
 
Text mining
Text miningText mining
Text mining
 
Spam Filtering
Spam FilteringSpam Filtering
Spam Filtering
 
Data Classification Presentation
Data Classification PresentationData Classification Presentation
Data Classification Presentation
 
Introduction to Text Mining
Introduction to Text MiningIntroduction to Text Mining
Introduction to Text Mining
 
Textmining Introduction
Textmining IntroductionTextmining Introduction
Textmining Introduction
 
Introduction to Sentiment Analysis
Introduction to Sentiment AnalysisIntroduction to Sentiment Analysis
Introduction to Sentiment Analysis
 

More from Albert Bifet

Artificial intelligence and data stream mining
Artificial intelligence and data stream miningArtificial intelligence and data stream mining
Artificial intelligence and data stream miningAlbert Bifet
 
MOA for the IoT at ACML 2016
MOA for the IoT at ACML 2016 MOA for the IoT at ACML 2016
MOA for the IoT at ACML 2016 Albert Bifet
 
Mining Big Data Streams with APACHE SAMOA
Mining Big Data Streams with APACHE SAMOAMining Big Data Streams with APACHE SAMOA
Mining Big Data Streams with APACHE SAMOAAlbert Bifet
 
Apache Samoa: Mining Big Data Streams with Apache Flink
Apache Samoa: Mining Big Data Streams with Apache FlinkApache Samoa: Mining Big Data Streams with Apache Flink
Apache Samoa: Mining Big Data Streams with Apache FlinkAlbert Bifet
 
Introduction to Big Data Science
Introduction to Big Data ScienceIntroduction to Big Data Science
Introduction to Big Data ScienceAlbert Bifet
 
Introduction to Big Data
Introduction to Big DataIntroduction to Big Data
Introduction to Big DataAlbert Bifet
 
Internet of Things Data Science
Internet of Things Data ScienceInternet of Things Data Science
Internet of Things Data ScienceAlbert Bifet
 
Real Time Big Data Management
Real Time Big Data ManagementReal Time Big Data Management
Real Time Big Data ManagementAlbert Bifet
 
A Short Course in Data Stream Mining
A Short Course in Data Stream MiningA Short Course in Data Stream Mining
A Short Course in Data Stream MiningAlbert Bifet
 
Real-Time Big Data Stream Analytics
Real-Time Big Data Stream AnalyticsReal-Time Big Data Stream Analytics
Real-Time Big Data Stream AnalyticsAlbert Bifet
 
Pitfalls in benchmarking data stream classification and how to avoid them
Pitfalls in benchmarking data stream classification and how to avoid themPitfalls in benchmarking data stream classification and how to avoid them
Pitfalls in benchmarking data stream classification and how to avoid themAlbert Bifet
 
STRIP: stream learning of influence probabilities.
STRIP: stream learning of influence probabilities.STRIP: stream learning of influence probabilities.
STRIP: stream learning of influence probabilities.Albert Bifet
 
Efficient Data Stream Classification via Probabilistic Adaptive Windows
Efficient Data Stream Classification via Probabilistic Adaptive WindowsEfficient Data Stream Classification via Probabilistic Adaptive Windows
Efficient Data Stream Classification via Probabilistic Adaptive WindowsAlbert Bifet
 
Mining Big Data in Real Time
Mining Big Data in Real TimeMining Big Data in Real Time
Mining Big Data in Real TimeAlbert Bifet
 
Mining Big Data in Real Time
Mining Big Data in Real TimeMining Big Data in Real Time
Mining Big Data in Real TimeAlbert Bifet
 
Mining Frequent Closed Graphs on Evolving Data Streams
Mining Frequent Closed Graphs on Evolving Data StreamsMining Frequent Closed Graphs on Evolving Data Streams
Mining Frequent Closed Graphs on Evolving Data StreamsAlbert Bifet
 
PAKDD 2011 TUTORIAL Handling Concept Drift: Importance, Challenges and Solutions
PAKDD 2011 TUTORIAL Handling Concept Drift: Importance, Challenges and SolutionsPAKDD 2011 TUTORIAL Handling Concept Drift: Importance, Challenges and Solutions
PAKDD 2011 TUTORIAL Handling Concept Drift: Importance, Challenges and SolutionsAlbert Bifet
 
Sentiment Knowledge Discovery in Twitter Streaming Data
Sentiment Knowledge Discovery in Twitter Streaming DataSentiment Knowledge Discovery in Twitter Streaming Data
Sentiment Knowledge Discovery in Twitter Streaming DataAlbert Bifet
 
Leveraging Bagging for Evolving Data Streams
Leveraging Bagging for Evolving Data StreamsLeveraging Bagging for Evolving Data Streams
Leveraging Bagging for Evolving Data StreamsAlbert Bifet
 
Fast Perceptron Decision Tree Learning from Evolving Data Streams
Fast Perceptron Decision Tree Learning from Evolving Data StreamsFast Perceptron Decision Tree Learning from Evolving Data Streams
Fast Perceptron Decision Tree Learning from Evolving Data StreamsAlbert Bifet
 

More from Albert Bifet (20)

Artificial intelligence and data stream mining
Artificial intelligence and data stream miningArtificial intelligence and data stream mining
Artificial intelligence and data stream mining
 
MOA for the IoT at ACML 2016
MOA for the IoT at ACML 2016 MOA for the IoT at ACML 2016
MOA for the IoT at ACML 2016
 
Mining Big Data Streams with APACHE SAMOA
Mining Big Data Streams with APACHE SAMOAMining Big Data Streams with APACHE SAMOA
Mining Big Data Streams with APACHE SAMOA
 
Apache Samoa: Mining Big Data Streams with Apache Flink
Apache Samoa: Mining Big Data Streams with Apache FlinkApache Samoa: Mining Big Data Streams with Apache Flink
Apache Samoa: Mining Big Data Streams with Apache Flink
 
Introduction to Big Data Science
Introduction to Big Data ScienceIntroduction to Big Data Science
Introduction to Big Data Science
 
Introduction to Big Data
Introduction to Big DataIntroduction to Big Data
Introduction to Big Data
 
Internet of Things Data Science
Internet of Things Data ScienceInternet of Things Data Science
Internet of Things Data Science
 
Real Time Big Data Management
Real Time Big Data ManagementReal Time Big Data Management
Real Time Big Data Management
 
A Short Course in Data Stream Mining
A Short Course in Data Stream MiningA Short Course in Data Stream Mining
A Short Course in Data Stream Mining
 
Real-Time Big Data Stream Analytics
Real-Time Big Data Stream AnalyticsReal-Time Big Data Stream Analytics
Real-Time Big Data Stream Analytics
 
Pitfalls in benchmarking data stream classification and how to avoid them
Pitfalls in benchmarking data stream classification and how to avoid themPitfalls in benchmarking data stream classification and how to avoid them
Pitfalls in benchmarking data stream classification and how to avoid them
 
STRIP: stream learning of influence probabilities.
STRIP: stream learning of influence probabilities.STRIP: stream learning of influence probabilities.
STRIP: stream learning of influence probabilities.
 
Efficient Data Stream Classification via Probabilistic Adaptive Windows
Efficient Data Stream Classification via Probabilistic Adaptive WindowsEfficient Data Stream Classification via Probabilistic Adaptive Windows
Efficient Data Stream Classification via Probabilistic Adaptive Windows
 
Mining Big Data in Real Time
Mining Big Data in Real TimeMining Big Data in Real Time
Mining Big Data in Real Time
 
Mining Big Data in Real Time
Mining Big Data in Real TimeMining Big Data in Real Time
Mining Big Data in Real Time
 
Mining Frequent Closed Graphs on Evolving Data Streams
Mining Frequent Closed Graphs on Evolving Data StreamsMining Frequent Closed Graphs on Evolving Data Streams
Mining Frequent Closed Graphs on Evolving Data Streams
 
PAKDD 2011 TUTORIAL Handling Concept Drift: Importance, Challenges and Solutions
PAKDD 2011 TUTORIAL Handling Concept Drift: Importance, Challenges and SolutionsPAKDD 2011 TUTORIAL Handling Concept Drift: Importance, Challenges and Solutions
PAKDD 2011 TUTORIAL Handling Concept Drift: Importance, Challenges and Solutions
 
Sentiment Knowledge Discovery in Twitter Streaming Data
Sentiment Knowledge Discovery in Twitter Streaming DataSentiment Knowledge Discovery in Twitter Streaming Data
Sentiment Knowledge Discovery in Twitter Streaming Data
 
Leveraging Bagging for Evolving Data Streams
Leveraging Bagging for Evolving Data StreamsLeveraging Bagging for Evolving Data Streams
Leveraging Bagging for Evolving Data Streams
 
Fast Perceptron Decision Tree Learning from Evolving Data Streams
Fast Perceptron Decision Tree Learning from Evolving Data StreamsFast Perceptron Decision Tree Learning from Evolving Data Streams
Fast Perceptron Decision Tree Learning from Evolving Data Streams
 

Recently uploaded

%in Soweto+277-882-255-28 abortion pills for sale in soweto
%in Soweto+277-882-255-28 abortion pills for sale in soweto%in Soweto+277-882-255-28 abortion pills for sale in soweto
%in Soweto+277-882-255-28 abortion pills for sale in sowetomasabamasaba
 
BUS PASS MANGEMENT SYSTEM USING PHP.pptx
BUS PASS MANGEMENT SYSTEM USING PHP.pptxBUS PASS MANGEMENT SYSTEM USING PHP.pptx
BUS PASS MANGEMENT SYSTEM USING PHP.pptxalwaysnagaraju26
 
%in Bahrain+277-882-255-28 abortion pills for sale in Bahrain
%in Bahrain+277-882-255-28 abortion pills for sale in Bahrain%in Bahrain+277-882-255-28 abortion pills for sale in Bahrain
%in Bahrain+277-882-255-28 abortion pills for sale in Bahrainmasabamasaba
 
%in Stilfontein+277-882-255-28 abortion pills for sale in Stilfontein
%in Stilfontein+277-882-255-28 abortion pills for sale in Stilfontein%in Stilfontein+277-882-255-28 abortion pills for sale in Stilfontein
%in Stilfontein+277-882-255-28 abortion pills for sale in Stilfonteinmasabamasaba
 
WSO2CON 2024 Slides - Unlocking Value with AI
WSO2CON 2024 Slides - Unlocking Value with AIWSO2CON 2024 Slides - Unlocking Value with AI
WSO2CON 2024 Slides - Unlocking Value with AIWSO2
 
%in Benoni+277-882-255-28 abortion pills for sale in Benoni
%in Benoni+277-882-255-28 abortion pills for sale in Benoni%in Benoni+277-882-255-28 abortion pills for sale in Benoni
%in Benoni+277-882-255-28 abortion pills for sale in Benonimasabamasaba
 
WSO2CON 2024 - Building the API First Enterprise – Running an API Program, fr...
WSO2CON 2024 - Building the API First Enterprise – Running an API Program, fr...WSO2CON 2024 - Building the API First Enterprise – Running an API Program, fr...
WSO2CON 2024 - Building the API First Enterprise – Running an API Program, fr...WSO2
 
Architecture decision records - How not to get lost in the past
Architecture decision records - How not to get lost in the pastArchitecture decision records - How not to get lost in the past
Architecture decision records - How not to get lost in the pastPapp Krisztián
 
Devoxx UK 2024 - Going serverless with Quarkus, GraalVM native images and AWS...
Devoxx UK 2024 - Going serverless with Quarkus, GraalVM native images and AWS...Devoxx UK 2024 - Going serverless with Quarkus, GraalVM native images and AWS...
Devoxx UK 2024 - Going serverless with Quarkus, GraalVM native images and AWS...Bert Jan Schrijver
 
VTU technical seminar 8Th Sem on Scikit-learn
VTU technical seminar 8Th Sem on Scikit-learnVTU technical seminar 8Th Sem on Scikit-learn
VTU technical seminar 8Th Sem on Scikit-learnAmarnathKambale
 
MarTech Trend 2024 Book : Marketing Technology Trends (2024 Edition) How Data...
MarTech Trend 2024 Book : Marketing Technology Trends (2024 Edition) How Data...MarTech Trend 2024 Book : Marketing Technology Trends (2024 Edition) How Data...
MarTech Trend 2024 Book : Marketing Technology Trends (2024 Edition) How Data...Jittipong Loespradit
 
%in ivory park+277-882-255-28 abortion pills for sale in ivory park
%in ivory park+277-882-255-28 abortion pills for sale in ivory park %in ivory park+277-882-255-28 abortion pills for sale in ivory park
%in ivory park+277-882-255-28 abortion pills for sale in ivory park masabamasaba
 
Announcing Codolex 2.0 from GDK Software
Announcing Codolex 2.0 from GDK SoftwareAnnouncing Codolex 2.0 from GDK Software
Announcing Codolex 2.0 from GDK SoftwareJim McKeeth
 
WSO2CON 2024 - Cloud Native Middleware: Domain-Driven Design, Cell-Based Arch...
WSO2CON 2024 - Cloud Native Middleware: Domain-Driven Design, Cell-Based Arch...WSO2CON 2024 - Cloud Native Middleware: Domain-Driven Design, Cell-Based Arch...
WSO2CON 2024 - Cloud Native Middleware: Domain-Driven Design, Cell-Based Arch...WSO2
 
%in Midrand+277-882-255-28 abortion pills for sale in midrand
%in Midrand+277-882-255-28 abortion pills for sale in midrand%in Midrand+277-882-255-28 abortion pills for sale in midrand
%in Midrand+277-882-255-28 abortion pills for sale in midrandmasabamasaba
 
%in tembisa+277-882-255-28 abortion pills for sale in tembisa
%in tembisa+277-882-255-28 abortion pills for sale in tembisa%in tembisa+277-882-255-28 abortion pills for sale in tembisa
%in tembisa+277-882-255-28 abortion pills for sale in tembisamasabamasaba
 
WSO2CON 2024 - Navigating API Complexity: REST, GraphQL, gRPC, Websocket, Web...
WSO2CON 2024 - Navigating API Complexity: REST, GraphQL, gRPC, Websocket, Web...WSO2CON 2024 - Navigating API Complexity: REST, GraphQL, gRPC, Websocket, Web...
WSO2CON 2024 - Navigating API Complexity: REST, GraphQL, gRPC, Websocket, Web...WSO2
 
%+27788225528 love spells in Boston Psychic Readings, Attraction spells,Bring...
%+27788225528 love spells in Boston Psychic Readings, Attraction spells,Bring...%+27788225528 love spells in Boston Psychic Readings, Attraction spells,Bring...
%+27788225528 love spells in Boston Psychic Readings, Attraction spells,Bring...masabamasaba
 

Recently uploaded (20)

%in Soweto+277-882-255-28 abortion pills for sale in soweto
%in Soweto+277-882-255-28 abortion pills for sale in soweto%in Soweto+277-882-255-28 abortion pills for sale in soweto
%in Soweto+277-882-255-28 abortion pills for sale in soweto
 
Abortion Pill Prices Tembisa [(+27832195400*)] 🏥 Women's Abortion Clinic in T...
Abortion Pill Prices Tembisa [(+27832195400*)] 🏥 Women's Abortion Clinic in T...Abortion Pill Prices Tembisa [(+27832195400*)] 🏥 Women's Abortion Clinic in T...
Abortion Pill Prices Tembisa [(+27832195400*)] 🏥 Women's Abortion Clinic in T...
 
BUS PASS MANGEMENT SYSTEM USING PHP.pptx
BUS PASS MANGEMENT SYSTEM USING PHP.pptxBUS PASS MANGEMENT SYSTEM USING PHP.pptx
BUS PASS MANGEMENT SYSTEM USING PHP.pptx
 
%in Bahrain+277-882-255-28 abortion pills for sale in Bahrain
%in Bahrain+277-882-255-28 abortion pills for sale in Bahrain%in Bahrain+277-882-255-28 abortion pills for sale in Bahrain
%in Bahrain+277-882-255-28 abortion pills for sale in Bahrain
 
%in Stilfontein+277-882-255-28 abortion pills for sale in Stilfontein
%in Stilfontein+277-882-255-28 abortion pills for sale in Stilfontein%in Stilfontein+277-882-255-28 abortion pills for sale in Stilfontein
%in Stilfontein+277-882-255-28 abortion pills for sale in Stilfontein
 
WSO2CON 2024 Slides - Unlocking Value with AI
WSO2CON 2024 Slides - Unlocking Value with AIWSO2CON 2024 Slides - Unlocking Value with AI
WSO2CON 2024 Slides - Unlocking Value with AI
 
%in Benoni+277-882-255-28 abortion pills for sale in Benoni
%in Benoni+277-882-255-28 abortion pills for sale in Benoni%in Benoni+277-882-255-28 abortion pills for sale in Benoni
%in Benoni+277-882-255-28 abortion pills for sale in Benoni
 
WSO2CON 2024 - Building the API First Enterprise – Running an API Program, fr...
WSO2CON 2024 - Building the API First Enterprise – Running an API Program, fr...WSO2CON 2024 - Building the API First Enterprise – Running an API Program, fr...
WSO2CON 2024 - Building the API First Enterprise – Running an API Program, fr...
 
Architecture decision records - How not to get lost in the past
Architecture decision records - How not to get lost in the pastArchitecture decision records - How not to get lost in the past
Architecture decision records - How not to get lost in the past
 
Devoxx UK 2024 - Going serverless with Quarkus, GraalVM native images and AWS...
Devoxx UK 2024 - Going serverless with Quarkus, GraalVM native images and AWS...Devoxx UK 2024 - Going serverless with Quarkus, GraalVM native images and AWS...
Devoxx UK 2024 - Going serverless with Quarkus, GraalVM native images and AWS...
 
VTU technical seminar 8Th Sem on Scikit-learn
VTU technical seminar 8Th Sem on Scikit-learnVTU technical seminar 8Th Sem on Scikit-learn
VTU technical seminar 8Th Sem on Scikit-learn
 
MarTech Trend 2024 Book : Marketing Technology Trends (2024 Edition) How Data...
MarTech Trend 2024 Book : Marketing Technology Trends (2024 Edition) How Data...MarTech Trend 2024 Book : Marketing Technology Trends (2024 Edition) How Data...
MarTech Trend 2024 Book : Marketing Technology Trends (2024 Edition) How Data...
 
%in ivory park+277-882-255-28 abortion pills for sale in ivory park
%in ivory park+277-882-255-28 abortion pills for sale in ivory park %in ivory park+277-882-255-28 abortion pills for sale in ivory park
%in ivory park+277-882-255-28 abortion pills for sale in ivory park
 
Announcing Codolex 2.0 from GDK Software
Announcing Codolex 2.0 from GDK SoftwareAnnouncing Codolex 2.0 from GDK Software
Announcing Codolex 2.0 from GDK Software
 
WSO2CON 2024 - Cloud Native Middleware: Domain-Driven Design, Cell-Based Arch...
WSO2CON 2024 - Cloud Native Middleware: Domain-Driven Design, Cell-Based Arch...WSO2CON 2024 - Cloud Native Middleware: Domain-Driven Design, Cell-Based Arch...
WSO2CON 2024 - Cloud Native Middleware: Domain-Driven Design, Cell-Based Arch...
 
%in Midrand+277-882-255-28 abortion pills for sale in midrand
%in Midrand+277-882-255-28 abortion pills for sale in midrand%in Midrand+277-882-255-28 abortion pills for sale in midrand
%in Midrand+277-882-255-28 abortion pills for sale in midrand
 
%in tembisa+277-882-255-28 abortion pills for sale in tembisa
%in tembisa+277-882-255-28 abortion pills for sale in tembisa%in tembisa+277-882-255-28 abortion pills for sale in tembisa
%in tembisa+277-882-255-28 abortion pills for sale in tembisa
 
WSO2CON 2024 - Navigating API Complexity: REST, GraphQL, gRPC, Websocket, Web...
WSO2CON 2024 - Navigating API Complexity: REST, GraphQL, gRPC, Websocket, Web...WSO2CON 2024 - Navigating API Complexity: REST, GraphQL, gRPC, Websocket, Web...
WSO2CON 2024 - Navigating API Complexity: REST, GraphQL, gRPC, Websocket, Web...
 
%+27788225528 love spells in Boston Psychic Readings, Attraction spells,Bring...
%+27788225528 love spells in Boston Psychic Readings, Attraction spells,Bring...%+27788225528 love spells in Boston Psychic Readings, Attraction spells,Bring...
%+27788225528 love spells in Boston Psychic Readings, Attraction spells,Bring...
 
Abortion Pills In Pretoria ](+27832195400*)[ 🏥 Women's Abortion Clinic In Pre...
Abortion Pills In Pretoria ](+27832195400*)[ 🏥 Women's Abortion Clinic In Pre...Abortion Pills In Pretoria ](+27832195400*)[ 🏥 Women's Abortion Clinic In Pre...
Abortion Pills In Pretoria ](+27832195400*)[ 🏥 Women's Abortion Clinic In Pre...
 

Multi-label Classification with Meta-labels

  • 1. Multi-Label Classification with Meta Labels Jesse Read(1), Antti Puurula(2), Albert Bifet(3) (1) Aalto University and HIIT, Finland (2) University of Waikato, New Zealand (3) Huawei Noah’s Ark Lab, Hong Kong 17 December 2014
  • 2. Overview Multi-label classification: ⊆ {beach, sunset, foliage, field, mountain, urban} • Most multi-label classification methods can be expressed in a general framework of meta-label classification • Our work combines labels into meta-labels, so as to learn dependence efficiently and effectively.
  • 3. Multi-label Learning With input variables X, produce predictions for multiple output variables Y . The basic binary relevance approach, YCYBYA X5X4X3X2X1 • does not capture label dependence (among Y -variables) • does not scale to large number of labels The label powerset approach models label combinations as class values in a multi-class problem.
  • 4. Meta Labels We introduce a layer of meta-labels, YCYBYA YB,CYA,B X5X4X3X2X1 • captures label dependence • meta-labels can be fewer than the original number of labels • and are deterministically decodable into the labels
  • 5. Producing Meta Labels dataset instance labels 1 B 2 B,C 3 C 4 B 5 A,C 6 A,C 7 A,C 8 A,B,C 9 C binary relevance YA YB YC 0 1 0 0 1 1 0 0 1 0 1 0 1 0 1 1 0 1 1 0 1 1 1 1 0 0 1 label powerset YA,B,C B BC C B AC AC AC ABC C meta labels YA,C YB,C ∅ B ∅ BC ∅ C ∅ B AC C AC C AC C ∅ BC ∅ C • binary relevance: 9 exs, 3 × 2 binary classes • label powerset: 9 exs, 1 × 5 multi-class
  • 6. Producing Meta Labels dataset instance labels 1 B 2 B,C 3 C 4 B 5 A,C 6 A,C 7 A,C 8 A,B,C 9 C binary relevance YA YB YC 0 1 0 0 1 1 0 0 1 0 1 0 1 0 1 1 0 1 1 0 1 1 1 1 0 0 1 label powerset YA,B,C B BC C B AC AC AC ABC C meta labels YA,C YB,C ∅ B ∅ BC ∅ C ∅ B AC C AC C AC C ∅ BC ∅ C • binary relevance: 9 exs, 3 × 2 binary classes • label powerset: 9 exs, 1 × 5 multi-class • pruned meta labels: 9 exs, 2 meta labels, of 2 and 3 values YAC ∈ {∅, AC}, YBC ∈ {B, C, BC} (one possible formulation)
  • 7. Example 2 There is no need to model labels together if there is no strong dependence between them, YCYBYA YCYA,B X1X2X3X4X5 YA,B ∈ {∅, B, AB}, YC ∈ {∅, C} (e.g., no strong relation between C and the other labels)
  • 8. General process for classification with meta-labels 1 Make a partition (either overlapping or disjoint) of the label set 2 Relabel the meta-labels, deciding on how many values each label can take (i.e., possibly pruning some) 3 Train classifiers to predict meta-labels from the input instances 4 Make predictions into the meta-label space 5 Recombine predictions into the label space
  • 9. Voting Using Meta Labels Table : Meta-label Vote, e.g., for YAB ∈ {∅, B, AB} v A B p(YAB = v|˜x) ∅ 0 0 0.0 B 0 1 0.9 AB 1 1 0.1 PAB 0.1 1.0 Table : Labelset Voting: From Meta-labels to Labels A B C PAB 0.1 1.0 PBC 0.7 0.3 k Pk,j 0.1 1.7 0.3 ˆy (> 0.5) 0 1 0
  • 10. Results 0 5 10 15 20 25 30 35 400.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 Accuracy RAkEL RAkELp 0 5 10 15 20 25 30 35 400 500 1000 1500 2000 Runningtime RAkEL RAkELp Figure : On Enron (1700 emails, 53 labels). RAkEL: random ensembles of ‘label-powerset method’ on subsets of size k (horizontal axis) vs RAkELp: with pruned meta-labels
  • 11. Summary • General framework of meta-labels for multi-label classification • Unifies various approaches from the literature • New models RAkELp and EpRd have large improvement in running time • Part of solution for LSHTC4 (1st place) and WISE (2nd Place) Kaggle challenges • Code available at http://meka.sourceforge.net
  • 12. Multi-Label Classification with Meta Labels Jesse Read(1), Antti Puurula(2), Albert Bifet(3) (1) Aalto University and HIIT, Finland (2) University of Waikato, New Zealand (3) Huawei Noah’s Ark Lab, Hong Kong 17 December 2014