The Benefits of a Model of Annotation

Size: px
Start display at page:

Download "The Benefits of a Model of Annotation"

Transcription

1 The Benefits of a Model of Annotation Rebecca J. Passonneau and Bob Carpenter Columbia University Center for Computational Learning Systems Department of Statistics LAW VII, August 2013

2 Conventional Approach To Corpus Quality Create gold-standard corpus A = B C B C Every instance in B labeled by all (2 to 4) annotators for interannotator reliability (consistency) Every instance in C labeled by a single annotator for ground truth dataset Assumptions There is only one way to be accurate If annotators are consistent, they must be accurate Corpus quality is good enough if annotators are consistent on B LAW VII 1

3 Word Sense Judgments: Adjectival Fair 6 of 10 WordNet senses used, 6 annotators Sometimes even many annotators agree 100% To be fair, I want to pass on to you a complaint that I do find valid, so you can better judge the situation. {sense1, sense1, sense1, sense1, sense1, sense1} High praise is tidy middling and middling is very fair! {sense5, sense5, sense5, sense5, sense5, sense5} What is the true label if annotators do not agree? And our ideas of what constitutes a fair wage or a fair return on capital are historically contingent. {sense1, sense1, sense1, sense2, sense2, sense2}... the federal government... is wrangling for its fair share of the dividend. {sense1, sense1, sense2,sense2, sense8, sense8} LAW VII 2

4 Problems Agreement measures consistency, not quality There is no way to distinguish > average from < average annotators B may not be a representative sample No information about quality of individual instances No method to infer ground truth label on instances where annotators do not agree LAW VII 3

5 Outline Related Work Limitations of Agreement Measures Probabilistic Model MASC word sense corpus Crowdsourced version Model results Class prevalence Annotator accuracy and bias Ground truth labels LAW VII 4

6 Related Work Dawid and Skene, 1979; Albert and Dodd, 2008; Smyth, 1995 Bruce and Wiebe, 1999; Snow et al., 2008 Rzhetsky et al., 2009; Whitehill et al., 2009 Hovy et al., 2013 (NAACL) LAW VII 5

7 Pairwise Agreement A [0, 1] Considers I Items, J Annotators Sum all pairs of annotators who agree on the label Normalize by the total number of annotator pairs No reference to rate of labels k 1 : K LAW VII 6

8 Agreement Coefficients IA [ 1, 1] Considers I Items, J Annotators, K Label Classes Chance-adjusted IA : A C 1 C C based on label proportions ψ m,k for each m J For annotators m, n, m n Kappa: C m,n = K k=1 ψ m,k ψ n,k Alpha: C m,n = K k=1 ψ2 k LAW VII 7

9 Drawbacks to Agreement Metrics 1. Intrinsically pairwise 2. Agreement on error indistinguishable from correct agreement 3. When chance agreement is high, chance-adjusted agreement is low (high prevalence categories) 4. Annotators can have identical bias on a category 5. Item-level effects (difficulty) can inflate agreement-in-error 6. Decision boundaries for agreement quality are arbitrary 7. Confidence intervals (rarely computed) can be wide enough to cross decision boundaries LAW VII 8

10 MASC Word Sense Sentence Corpus 116 lemmas: 29 adjectives, 46 nouns, 41 verbs 1,000 example sentences per lemma drawn from the MASC corpus, a heterogeneous corpus of 19 genres 2,392,873 words counting every sentence once 3,328,815 words counting every sentence once for each annotated word 7.2 WordNet senses per word LAW VII 9

11 Word Sense Annotation Procedures Annotators College students from Vassar, Columbia, Barnard Trained using guidelines from Christiane Fellbaum SATANiC Graphical User Interface, with SVN Procedures Annotation rounds of 10 words per round, 2 to 4 annotators Pre-annotation sample: 50 sentences, to learn and review sense inventory 900 sentences annotated with one annotator per sentence 100 sentences annotated by all annotators for agreement LAW VII 10

12 Agreement Results Go from High to Low Word Pos Senses α Agreement late adj high adj severe adj strike noun date noun success noun mature verb add verb ask verb One label per instance may not be enough Model-based annotation evaluation can improve results LAW VII 11

13 Crowdsourced Word Sense Annotation Amazon Mechanical Turk 45 of the 116 words Same sentences, 20 to 25 labels per sentence 1 HIT: 10 sentences (100 HITs per word) Extensive piloting of HIT design/pricing/qualifications 90% lifetime approval rating 20,000 approved hits U.S. domain 228 turkers LAW VII 12

14 Sample HIT LAW VII 13

15 Data Format Each of the n observed labels is a tuple of an item ii I an annotator jj J a label y K n ii n jj n y n LAW VII 14

16 Dawid and Skene Model Joint probability of: true labels z i 1 : K; z Categorical(π) prevalence π k K-simplex annotator probability θ j,k,k of assigning k when true label is k (also a simplex, i.e., probabilities must be 0 and sum to 1) observed labels y n Categorical(θ jj[n],z[ii[n]] ) annotator jj[n] s responses to items ii[n] where true category is zz[ii[n]] LAW VII 15

17 Inference Additively smoothed Maximum Likelihood Estimation Equivalent to maximum a posteriori estimation in a Bayesian model with Dirichlet priors θ j,k Dirichet(α k ) π Dirichlet(β) LAW VII 16

18 Prevalence: Add (v); α = 0.55; A = add-v MASC Freq MASC Maj MASC MLE AMT Maj AMT MLE 0.00 Other Sense 1 Sense 2 Sense 3 Sense 4 Sense 5 Sense 6 LAW VII 17

19 Prevalence: Help (v); α = 0.26; A = help-v MASC Freq MASC Maj MASC MLE AMT Maj AMT MLE Other Sense 1 Sense 2 Sense 3 Sense 4 Sense 5 Sense 6 Sense 7 Sense 8 LAW VII 18

20 Annotator Heatmaps Add (v); α = 0.55; A = 0.72 LAW VII 19

21 Annotator Heatmaps Help (v); α = 0.26; A = 0.58 LAW VII 20

22 Ground Truth Labels Add (verb); α = 0.55; A = 0.72 Sense k 0.99 Prop make an addition; join or state or say further bestow a quality on constitute an addition SubTot Rest LAW VII 21

23 Ground Truth Labels Help (verb); α = 0.26; A = 0.58 Sense k 0.99 Prop give assistance; be of service improve the condition of be of use contribute to the furtherance of SubTot Rest LAW VII 22

24 Cost Comparison MASC AMT Lemmas Creating infrastructure 1 year 0.05 year Annotation period 5 years 0.05 year Cost of annotators $80,000 $15,000 Cost per ground truth label $0.70 $0.33 LAW VII 23

25 Summary of Contributions More information about the annotated data Higher quality corpus Lower cost Valuable corpus on moderately fine-grained word sense LAW VII 24

26 Future Work Richer models, e.g., add a parameter for item difficulty Annotate more of the 71 remaining MASC lemmas Investigate/monitor utility of the corpus to Train WSD Study WordNet sense inventories Evaluate WSD using probability distribution of senses per item Develop overall quality measure of the corpus (entropy) LAW VII 25

Statistical Models of the Annotation Process

Statistical Models of the Annotation Process Bob Carpenter 1 Massimo Poesio 2 1 Alias-I 2 Università di Trento LREC 2010 Tutorial 17th May 2010 Many slides due to Ron Artstein Annotated corpora Annotated corpora are needed for: Supervised learning

More information

Crowdsourcing via Tensor Augmentation and Completion (TAC)

Crowdsourcing via Tensor Augmentation and Completion (TAC) Crowdsourcing via Tensor Augmentation and Completion (TAC) Presenter: Yao Zhou joint work with: Dr. Jingrui He - 1 - Roadmap Background Related work Crowdsourcing based on TAC Experimental results Conclusion

More information

Aggregating Ordinal Labels from Crowds by Minimax Conditional Entropy. Denny Zhou Qiang Liu John Platt Chris Meek

Aggregating Ordinal Labels from Crowds by Minimax Conditional Entropy. Denny Zhou Qiang Liu John Platt Chris Meek Aggregating Ordinal Labels from Crowds by Minimax Conditional Entropy Denny Zhou Qiang Liu John Platt Chris Meek 2 Crowds vs experts labeling: strength Time saving Money saving Big labeled data More data

More information

Parametric Models. Dr. Shuang LIANG. School of Software Engineering TongJi University Fall, 2012

Parametric Models. Dr. Shuang LIANG. School of Software Engineering TongJi University Fall, 2012 Parametric Models Dr. Shuang LIANG School of Software Engineering TongJi University Fall, 2012 Today s Topics Maximum Likelihood Estimation Bayesian Density Estimation Today s Topics Maximum Likelihood

More information

Learning from the Wisdom of Crowds by Minimax Entropy. Denny Zhou, John Platt, Sumit Basu and Yi Mao Microsoft Research, Redmond, WA

Learning from the Wisdom of Crowds by Minimax Entropy. Denny Zhou, John Platt, Sumit Basu and Yi Mao Microsoft Research, Redmond, WA Learning from the Wisdom of Crowds by Minimax Entropy Denny Zhou, John Platt, Sumit Basu and Yi Mao Microsoft Research, Redmond, WA Outline 1. Introduction 2. Minimax entropy principle 3. Future work and

More information

Efficient Online Scalar Annotation with Bounded Support

Efficient Online Scalar Annotation with Bounded Support aaab7xicbvbns8naej3ur1q/qh69lbbbiyurqbvvxisygyhdwwz3brln5uwoxfk6i/w4khfq//hm//gbzudtj4yelw3w8y8mjxcoot+o6wv1bx1jfjmzwt7z3evun/wajjmm+6zrca6hvldpvdcr4gst1pnarxk3gpht1o/9cs1eyl6whhkg5goligeoil1llxqajhvwrnrbszkgxifaqgbzq96le3n7as5gqzpmzpdffikcabzn8uulmhqeujeiadyxvnoymygfntsijvfoksrqthwsm/p7iawzmoa5tzxxaba9qfif18kwugpyodimuwlzrvemcszk+jvpc8zyrellglhbyvssdvlaboqbc8xzexix9ev6679xe1xkrrhmo4bhowynlamadnmehbin4hld4c1lnxxl3puatjaeyoyq/cd5/ajgyjy=

More information

Hidden Markov Models in Language Processing

Hidden Markov Models in Language Processing Hidden Markov Models in Language Processing Dustin Hillard Lecture notes courtesy of Prof. Mari Ostendorf Outline Review of Markov models What is an HMM? Examples General idea of hidden variables: implications

More information

Deconstructing Data Science

Deconstructing Data Science Deconstructing Data Science David Bamman, UC Berkeley Info 290 Lecture 3: Classification overview Jan 24, 2017 Auditors Send me an email to get access to bcourses (announcements, readings, etc.) Classification

More information

Prenominal Modifier Ordering via MSA. Alignment

Prenominal Modifier Ordering via MSA. Alignment Introduction Prenominal Modifier Ordering via Multiple Sequence Alignment Aaron Dunlop Margaret Mitchell 2 Brian Roark Oregon Health & Science University Portland, OR 2 University of Aberdeen Aberdeen,

More information

Evaluation Strategies

Evaluation Strategies Evaluation Intrinsic Evaluation Comparison with an ideal output: Challenges: Requires a large testing set Intrinsic subjectivity of some discourse related judgments Hard to find corpora for training/testing

More information

ECE521 week 3: 23/26 January 2017

ECE521 week 3: 23/26 January 2017 ECE521 week 3: 23/26 January 2017 Outline Probabilistic interpretation of linear regression - Maximum likelihood estimation (MLE) - Maximum a posteriori (MAP) estimation Bias-variance trade-off Linear

More information

Text Mining. Dr. Yanjun Li. Associate Professor. Department of Computer and Information Sciences Fordham University

Text Mining. Dr. Yanjun Li. Associate Professor. Department of Computer and Information Sciences Fordham University Text Mining Dr. Yanjun Li Associate Professor Department of Computer and Information Sciences Fordham University Outline Introduction: Data Mining Part One: Text Mining Part Two: Preprocessing Text Data

More information

Intelligent Systems:

Intelligent Systems: Intelligent Systems: Undirected Graphical models (Factor Graphs) (2 lectures) Carsten Rother 15/01/2015 Intelligent Systems: Probabilistic Inference in DGM and UGM Roadmap for next two lectures Definition

More information

10 : HMM and CRF. 1 Case Study: Supervised Part-of-Speech Tagging

10 : HMM and CRF. 1 Case Study: Supervised Part-of-Speech Tagging 10-708: Probabilistic Graphical Models 10-708, Spring 2018 10 : HMM and CRF Lecturer: Kayhan Batmanghelich Scribes: Ben Lengerich, Michael Kleyman 1 Case Study: Supervised Part-of-Speech Tagging We will

More information

Part of Speech Tagging: Viterbi, Forward, Backward, Forward- Backward, Baum-Welch. COMP-599 Oct 1, 2015

Part of Speech Tagging: Viterbi, Forward, Backward, Forward- Backward, Baum-Welch. COMP-599 Oct 1, 2015 Part of Speech Tagging: Viterbi, Forward, Backward, Forward- Backward, Baum-Welch COMP-599 Oct 1, 2015 Announcements Research skills workshop today 3pm-4:30pm Schulich Library room 313 Start thinking about

More information

A brief introduction to Conditional Random Fields

A brief introduction to Conditional Random Fields A brief introduction to Conditional Random Fields Mark Johnson Macquarie University April, 2005, updated October 2010 1 Talk outline Graphical models Maximum likelihood and maximum conditional likelihood

More information

Learning Classification with Auxiliary Probabilistic Information Quang Nguyen Hamed Valizadegan Milos Hauskrecht

Learning Classification with Auxiliary Probabilistic Information Quang Nguyen Hamed Valizadegan Milos Hauskrecht Learning Classification with Auxiliary Probabilistic Information Quang Nguyen Hamed Valizadegan Milos Hauskrecht Computer Science Department University of Pittsburgh Outline Introduction Learning with

More information

Maximum Likelihood Estimation. only training data is available to design a classifier

Maximum Likelihood Estimation. only training data is available to design a classifier Introduction to Pattern Recognition [ Part 5 ] Mahdi Vasighi Introduction Bayesian Decision Theory shows that we could design an optimal classifier if we knew: P( i ) : priors p(x i ) : class-conditional

More information

Bayesian Models in Machine Learning

Bayesian Models in Machine Learning Bayesian Models in Machine Learning Lukáš Burget Escuela de Ciencias Informáticas 2017 Buenos Aires, July 24-29 2017 Frequentist vs. Bayesian Frequentist point of view: Probability is the frequency of

More information

Preference-Based Rank Elicitation using Statistical Models: The Case of Mallows

Preference-Based Rank Elicitation using Statistical Models: The Case of Mallows Preference-Based Rank Elicitation using Statistical Models: The Case of Mallows Robert Busa-Fekete 1 Eyke Huellermeier 2 Balzs Szrnyi 3 1 MTA-SZTE Research Group on Artificial Intelligence, Tisza Lajos

More information

Learning Bayesian network : Given structure and completely observed data

Learning Bayesian network : Given structure and completely observed data Learning Bayesian network : Given structure and completely observed data Probabilistic Graphical Models Sharif University of Technology Spring 2017 Soleymani Learning problem Target: true distribution

More information

Maximum Entropy Markov Models

Maximum Entropy Markov Models Wi nøt trei a høliday in Sweden this yër? September 19th 26 Background Preliminary CRF work. Published in 2. Authors: McCallum, Freitag and Pereira. Key concepts: Maximum entropy / Overlapping features.

More information

Intelligent Systems (AI-2)

Intelligent Systems (AI-2) Intelligent Systems (AI-2) Computer Science cpsc422, Lecture 19 Oct, 24, 2016 Slide Sources Raymond J. Mooney University of Texas at Austin D. Koller, Stanford CS - Probabilistic Graphical Models D. Page,

More information

DEPARTMENT OF COMPUTER SCIENCE Autumn Semester MACHINE LEARNING AND ADAPTIVE INTELLIGENCE

DEPARTMENT OF COMPUTER SCIENCE Autumn Semester MACHINE LEARNING AND ADAPTIVE INTELLIGENCE Data Provided: None DEPARTMENT OF COMPUTER SCIENCE Autumn Semester 203 204 MACHINE LEARNING AND ADAPTIVE INTELLIGENCE 2 hours Answer THREE of the four questions. All questions carry equal weight. Figures

More information

The Naïve Bayes Classifier. Machine Learning Fall 2017

The Naïve Bayes Classifier. Machine Learning Fall 2017 The Naïve Bayes Classifier Machine Learning Fall 2017 1 Today s lecture The naïve Bayes Classifier Learning the naïve Bayes Classifier Practical concerns 2 Today s lecture The naïve Bayes Classifier Learning

More information

Mining Subjective Properties on the Web

Mining Subjective Properties on the Web Mining Subjective Properties on the Web Immanuel Trummer EPFL Lausanne, Switzerland immanuel.trummer@epfl.ch Sunita Sarawagi Google, Inc. and IIT Bombay Mountain View, USA/Mumbai, India sarawagi@google.com

More information

Distributed ML for DOSNs: giving power back to users

Distributed ML for DOSNs: giving power back to users Distributed ML for DOSNs: giving power back to users Amira Soliman KTH isocial Marie Curie Initial Training Networks Part1 Agenda DOSNs and Machine Learning DIVa: Decentralized Identity Validation for

More information

Classical and Bayesian inference

Classical and Bayesian inference Classical and Bayesian inference AMS 132 Claudia Wehrhahn (UCSC) Classical and Bayesian inference January 8 1 / 8 Probability and Statistical Models Motivating ideas AMS 131: Suppose that the random variable

More information

Bayesian Methods: Naïve Bayes

Bayesian Methods: Naïve Bayes Bayesian Methods: aïve Bayes icholas Ruozzi University of Texas at Dallas based on the slides of Vibhav Gogate Last Time Parameter learning Learning the parameter of a simple coin flipping model Prior

More information

Uncovering the Latent Structures of Crowd Labeling

Uncovering the Latent Structures of Crowd Labeling Uncovering the Latent Structures of Crowd Labeling Tian Tian and Jun Zhu Presenter:XXX Tsinghua University 1 / 26 Motivation Outline 1 Motivation 2 Related Works 3 Crowdsourcing Latent Class 4 Experiments

More information

Data Mining Part 4. Prediction

Data Mining Part 4. Prediction Data Mining Part 4. Prediction 4.3. Fall 2009 Instructor: Dr. Masoud Yaghini Outline Introduction Bayes Theorem Naïve References Introduction Bayesian classifiers A statistical classifiers Introduction

More information

Generative Clustering, Topic Modeling, & Bayesian Inference

Generative Clustering, Topic Modeling, & Bayesian Inference Generative Clustering, Topic Modeling, & Bayesian Inference INFO-4604, Applied Machine Learning University of Colorado Boulder December 12-14, 2017 Prof. Michael Paul Unsupervised Naïve Bayes Last week

More information

Data Mining: Concepts and Techniques. (3 rd ed.) Chapter 8. Chapter 8. Classification: Basic Concepts

Data Mining: Concepts and Techniques. (3 rd ed.) Chapter 8. Chapter 8. Classification: Basic Concepts Data Mining: Concepts and Techniques (3 rd ed.) Chapter 8 Chapter 8. Classification: Basic Concepts Classification: Basic Concepts Decision Tree Induction Bayes Classification Methods Rule-Based Classification

More information

Hierarchical Models of Data Coding

Hierarchical Models of Data Coding Hierarchical Models of Data Coding Bob Carpenter (w. Emily Jamison and Breck Baldwin) Alias-i, Inc. Supervised Machine Learning 1. Define coding standard mapping inputs to outputs, e.g.: English word stem

More information

Language as a Stochastic Process

Language as a Stochastic Process CS769 Spring 2010 Advanced Natural Language Processing Language as a Stochastic Process Lecturer: Xiaojin Zhu jerryzhu@cs.wisc.edu 1 Basic Statistics for NLP Pick an arbitrary letter x at random from any

More information

Sequence Modelling with Features: Linear-Chain Conditional Random Fields. COMP-599 Oct 6, 2015

Sequence Modelling with Features: Linear-Chain Conditional Random Fields. COMP-599 Oct 6, 2015 Sequence Modelling with Features: Linear-Chain Conditional Random Fields COMP-599 Oct 6, 2015 Announcement A2 is out. Due Oct 20 at 1pm. 2 Outline Hidden Markov models: shortcomings Generative vs. discriminative

More information

Spectral Unsupervised Parsing with Additive Tree Metrics

Spectral Unsupervised Parsing with Additive Tree Metrics Spectral Unsupervised Parsing with Additive Tree Metrics Ankur Parikh, Shay Cohen, Eric P. Xing Carnegie Mellon, University of Edinburgh Ankur Parikh 2014 1 Overview Model: We present a novel approach

More information

Undirected Graphical Models

Undirected Graphical Models Outline Hong Chang Institute of Computing Technology, Chinese Academy of Sciences Machine Learning Methods (Fall 2012) Outline Outline I 1 Introduction 2 Properties Properties 3 Generative vs. Conditional

More information

Crowdsourcing Pareto-Optimal Object Finding by Pairwise Comparisons

Crowdsourcing Pareto-Optimal Object Finding by Pairwise Comparisons 2015 The University of Texas at Arlington. All Rights Reserved. Crowdsourcing Pareto-Optimal Object Finding by Pairwise Comparisons Abolfazl Asudeh, Gensheng Zhang, Naeemul Hassan, Chengkai Li, Gergely

More information

Intelligent Systems (AI-2)

Intelligent Systems (AI-2) Intelligent Systems (AI-2) Computer Science cpsc422, Lecture 19 Oct, 23, 2015 Slide Sources Raymond J. Mooney University of Texas at Austin D. Koller, Stanford CS - Probabilistic Graphical Models D. Page,

More information

Bowl Maximum Entropy #4 By Ejay Weiss. Maxent Models: Maximum Entropy Foundations. By Yanju Chen. A Basic Comprehension with Derivations

Bowl Maximum Entropy #4 By Ejay Weiss. Maxent Models: Maximum Entropy Foundations. By Yanju Chen. A Basic Comprehension with Derivations Bowl Maximum Entropy #4 By Ejay Weiss Maxent Models: Maximum Entropy Foundations By Yanju Chen A Basic Comprehension with Derivations Outlines Generative vs. Discriminative Feature-Based Models Softmax

More information

Conditional Random Fields and beyond DANIEL KHASHABI CS 546 UIUC, 2013

Conditional Random Fields and beyond DANIEL KHASHABI CS 546 UIUC, 2013 Conditional Random Fields and beyond DANIEL KHASHABI CS 546 UIUC, 2013 Outline Modeling Inference Training Applications Outline Modeling Problem definition Discriminative vs. Generative Chain CRF General

More information

Naïve Bayesian. From Han Kamber Pei

Naïve Bayesian. From Han Kamber Pei Naïve Bayesian From Han Kamber Pei Bayesian Theorem: Basics Let X be a data sample ( evidence ): class label is unknown Let H be a hypothesis that X belongs to class C Classification is to determine H

More information

Budget-Optimal Task Allocation for Reliable Crowdsourcing Systems

Budget-Optimal Task Allocation for Reliable Crowdsourcing Systems Budget-Optimal Task Allocation for Reliable Crowdsourcing Systems Sewoong Oh Massachusetts Institute of Technology joint work with David R. Karger and Devavrat Shah September 28, 2011 1 / 13 Crowdsourcing

More information

TnT Part of Speech Tagger

TnT Part of Speech Tagger TnT Part of Speech Tagger By Thorsten Brants Presented By Arghya Roy Chaudhuri Kevin Patel Satyam July 29, 2014 1 / 31 Outline 1 Why Then? Why Now? 2 Underlying Model Other technicalities 3 Evaluation

More information

UNIVERSITY of PENNSYLVANIA CIS 520: Machine Learning Final, Fall 2013

UNIVERSITY of PENNSYLVANIA CIS 520: Machine Learning Final, Fall 2013 UNIVERSITY of PENNSYLVANIA CIS 520: Machine Learning Final, Fall 2013 Exam policy: This exam allows two one-page, two-sided cheat sheets; No other materials. Time: 2 hours. Be sure to write your name and

More information

Partially Directed Graphs and Conditional Random Fields. Sargur Srihari

Partially Directed Graphs and Conditional Random Fields. Sargur Srihari Partially Directed Graphs and Conditional Random Fields Sargur srihari@cedar.buffalo.edu 1 Topics Conditional Random Fields Gibbs distribution and CRF Directed and Undirected Independencies View as combination

More information

Classification, Linear Models, Naïve Bayes

Classification, Linear Models, Naïve Bayes Classification, Linear Models, Naïve Bayes CMSC 470 Marine Carpuat Slides credit: Dan Jurafsky & James Martin, Jacob Eisenstein Today Text classification problems and their evaluation Linear classifiers

More information

Quantization. Robert M. Haralick. Computer Science, Graduate Center City University of New York

Quantization. Robert M. Haralick. Computer Science, Graduate Center City University of New York Quantization Robert M. Haralick Computer Science, Graduate Center City University of New York Outline Quantizing 1 Quantizing 2 3 4 5 6 Quantizing Data is real-valued Data is integer valued with large

More information

Gaussian Processes. Le Song. Machine Learning II: Advanced Topics CSE 8803ML, Spring 2012

Gaussian Processes. Le Song. Machine Learning II: Advanced Topics CSE 8803ML, Spring 2012 Gaussian Processes Le Song Machine Learning II: Advanced Topics CSE 8803ML, Spring 01 Pictorial view of embedding distribution Transform the entire distribution to expected features Feature space Feature

More information

Model verification / validation A distributions-oriented approach

Model verification / validation A distributions-oriented approach Model verification / validation A distributions-oriented approach Dr. Christian Ohlwein Hans-Ertel-Centre for Weather Research Meteorological Institute, University of Bonn, Germany Ringvorlesung: Quantitative

More information

Probabilistic Machine Learning. Industrial AI Lab.

Probabilistic Machine Learning. Industrial AI Lab. Probabilistic Machine Learning Industrial AI Lab. Probabilistic Linear Regression Outline Probabilistic Classification Probabilistic Clustering Probabilistic Dimension Reduction 2 Probabilistic Linear

More information

CS340 Winter 2010: HW3 Out Wed. 2nd February, due Friday 11th February

CS340 Winter 2010: HW3 Out Wed. 2nd February, due Friday 11th February CS340 Winter 2010: HW3 Out Wed. 2nd February, due Friday 11th February 1 PageRank You are given in the file adjency.mat a matrix G of size n n where n = 1000 such that { 1 if outbound link from i to j,

More information

Measuring Variability in Sentence Ordering for News Summarization

Measuring Variability in Sentence Ordering for News Summarization Measuring Variability in Sentence Ordering for News Summarization Nitin Madnani a,b Rebecca Passonneau c Necip Fazil Ayan a,b John M. Conroy d Bonnie J. Dorr a,b Judith L. Klavans e Dianne P. O Leary a,b

More information

Estimation of reliability parameters from Experimental data (Parte 2) Prof. Enrico Zio

Estimation of reliability parameters from Experimental data (Parte 2) Prof. Enrico Zio Estimation of reliability parameters from Experimental data (Parte 2) This lecture Life test (t 1,t 2,...,t n ) Estimate θ of f T t θ For example: λ of f T (t)= λe - λt Classical approach (frequentist

More information

CS6220: DATA MINING TECHNIQUES

CS6220: DATA MINING TECHNIQUES CS6220: DATA MINING TECHNIQUES Matrix Data: Classification: Part 2 Instructor: Yizhou Sun yzsun@ccs.neu.edu September 21, 2014 Methods to Learn Matrix Data Set Data Sequence Data Time Series Graph & Network

More information

Introduction to Machine Learning Midterm Exam

Introduction to Machine Learning Midterm Exam 10-701 Introduction to Machine Learning Midterm Exam Instructors: Eric Xing, Ziv Bar-Joseph 17 November, 2015 There are 11 questions, for a total of 100 points. This exam is open book, open notes, but

More information

Crowd-Learning: Improving the Quality of Crowdsourcing Using Sequential Learning

Crowd-Learning: Improving the Quality of Crowdsourcing Using Sequential Learning Crowd-Learning: Improving the Quality of Crowdsourcing Using Sequential Learning Mingyan Liu (Joint work with Yang Liu) Department of Electrical Engineering and Computer Science University of Michigan,

More information

Data Mining Chapter 4: Data Analysis and Uncertainty Fall 2011 Ming Li Department of Computer Science and Technology Nanjing University

Data Mining Chapter 4: Data Analysis and Uncertainty Fall 2011 Ming Li Department of Computer Science and Technology Nanjing University Data Mining Chapter 4: Data Analysis and Uncertainty Fall 2011 Ming Li Department of Computer Science and Technology Nanjing University Why uncertainty? Why should data mining care about uncertainty? We

More information

Bayesian Analysis for Natural Language Processing Lecture 2

Bayesian Analysis for Natural Language Processing Lecture 2 Bayesian Analysis for Natural Language Processing Lecture 2 Shay Cohen February 4, 2013 Administrativia The class has a mailing list: coms-e6998-11@cs.columbia.edu Need two volunteers for leading a discussion

More information

The Bayes classifier

The Bayes classifier The Bayes classifier Consider where is a random vector in is a random variable (depending on ) Let be a classifier with probability of error/risk given by The Bayes classifier (denoted ) is the optimal

More information

Potential Outcomes Model (POM)

Potential Outcomes Model (POM) Potential Outcomes Model (POM) Relationship Between Counterfactual States Causality Empirical Strategies in Labor Economics, Angrist Krueger (1999): The most challenging empirical questions in economics

More information

Probabilistic Graphical Models: MRFs and CRFs. CSE628: Natural Language Processing Guest Lecturer: Veselin Stoyanov

Probabilistic Graphical Models: MRFs and CRFs. CSE628: Natural Language Processing Guest Lecturer: Veselin Stoyanov Probabilistic Graphical Models: MRFs and CRFs CSE628: Natural Language Processing Guest Lecturer: Veselin Stoyanov Why PGMs? PGMs can model joint probabilities of many events. many techniques commonly

More information

Introduction to Machine Learning Midterm Exam Solutions

Introduction to Machine Learning Midterm Exam Solutions 10-701 Introduction to Machine Learning Midterm Exam Solutions Instructors: Eric Xing, Ziv Bar-Joseph 17 November, 2015 There are 11 questions, for a total of 100 points. This exam is open book, open notes,

More information

Bayesian Learning. HT2015: SC4 Statistical Data Mining and Machine Learning. Maximum Likelihood Principle. The Bayesian Learning Framework

Bayesian Learning. HT2015: SC4 Statistical Data Mining and Machine Learning. Maximum Likelihood Principle. The Bayesian Learning Framework HT5: SC4 Statistical Data Mining and Machine Learning Dino Sejdinovic Department of Statistics Oxford http://www.stats.ox.ac.uk/~sejdinov/sdmml.html Maximum Likelihood Principle A generative model for

More information

Hidden Markov Models

Hidden Markov Models CS769 Spring 2010 Advanced Natural Language Processing Hidden Markov Models Lecturer: Xiaojin Zhu jerryzhu@cs.wisc.edu 1 Part-of-Speech Tagging The goal of Part-of-Speech (POS) tagging is to label each

More information

Statistical methods for NLP Estimation

Statistical methods for NLP Estimation Statistical methods for NLP Estimation UNIVERSITY OF Richard Johansson January 29, 2015 why does the teacher care so much about the coin-tossing experiment? because it can model many situations: I pick

More information

Quadratic and Other Inequalities in One Variable

Quadratic and Other Inequalities in One Variable Quadratic and Other Inequalities in One Variable If a quadratic equation is not in the standard form equaling zero, but rather uses an inequality sign ( , ), the equation is said to be a quadratic

More information

Chapter 19. Agreement and the kappa statistic

Chapter 19. Agreement and the kappa statistic 19. Agreement Chapter 19 Agreement and the kappa statistic Besides the 2 2contingency table for unmatched data and the 2 2table for matched data, there is a third common occurrence of data appearing summarised

More information

Hybrid Models for Text and Graphs. 10/23/2012 Analysis of Social Media

Hybrid Models for Text and Graphs. 10/23/2012 Analysis of Social Media Hybrid Models for Text and Graphs 10/23/2012 Analysis of Social Media Newswire Text Formal Primary purpose: Inform typical reader about recent events Broad audience: Explicitly establish shared context

More information

Empirical Methods in Natural Language Processing Lecture 11 Part-of-speech tagging and HMMs

Empirical Methods in Natural Language Processing Lecture 11 Part-of-speech tagging and HMMs Empirical Methods in Natural Language Processing Lecture 11 Part-of-speech tagging and HMMs (based on slides by Sharon Goldwater and Philipp Koehn) 21 February 2018 Nathan Schneider ENLP Lecture 11 21

More information

Probabilistic modeling. The slides are closely adapted from Subhransu Maji s slides

Probabilistic modeling. The slides are closely adapted from Subhransu Maji s slides Probabilistic modeling The slides are closely adapted from Subhransu Maji s slides Overview So far the models and algorithms you have learned about are relatively disconnected Probabilistic modeling framework

More information

Bayesian Learning Features of Bayesian learning methods:

Bayesian Learning Features of Bayesian learning methods: Bayesian Learning Features of Bayesian learning methods: Each observed training example can incrementally decrease or increase the estimated probability that a hypothesis is correct. This provides a more

More information

CSCI 5832 Natural Language Processing. Today 2/19. Statistical Sequence Classification. Lecture 9

CSCI 5832 Natural Language Processing. Today 2/19. Statistical Sequence Classification. Lecture 9 CSCI 5832 Natural Language Processing Jim Martin Lecture 9 1 Today 2/19 Review HMMs for POS tagging Entropy intuition Statistical Sequence classifiers HMMs MaxEnt MEMMs 2 Statistical Sequence Classification

More information

Improving Quality of Crowdsourced Labels via Probabilistic Matrix Factorization

Improving Quality of Crowdsourced Labels via Probabilistic Matrix Factorization Human Computation AAAI Technical Report WS-12-08 Improving Quality of Crowdsourced Labels via Probabilistic Matrix Factorization Hyun Joon Jung School of Information University of Texas at Austin hyunjoon@utexas.edu

More information

INF4820: Algorithms for Artificial Intelligence and Natural Language Processing. Hidden Markov Models

INF4820: Algorithms for Artificial Intelligence and Natural Language Processing. Hidden Markov Models INF4820: Algorithms for Artificial Intelligence and Natural Language Processing Hidden Markov Models Murhaf Fares & Stephan Oepen Language Technology Group (LTG) October 27, 2016 Recap: Probabilistic Language

More information

Categorization ANLP Lecture 10 Text Categorization with Naive Bayes

Categorization ANLP Lecture 10 Text Categorization with Naive Bayes 1 Categorization ANLP Lecture 10 Text Categorization with Naive Bayes Sharon Goldwater 6 October 2014 Important task for both humans and machines object identification face recognition spoken word recognition

More information

ANLP Lecture 10 Text Categorization with Naive Bayes

ANLP Lecture 10 Text Categorization with Naive Bayes ANLP Lecture 10 Text Categorization with Naive Bayes Sharon Goldwater 6 October 2014 Categorization Important task for both humans and machines 1 object identification face recognition spoken word recognition

More information

Density Estimation: ML, MAP, Bayesian estimation

Density Estimation: ML, MAP, Bayesian estimation Density Estimation: ML, MAP, Bayesian estimation CE-725: Statistical Pattern Recognition Sharif University of Technology Spring 2013 Soleymani Outline Introduction Maximum-Likelihood Estimation Maximum

More information

RETRIEVAL MODELS. Dr. Gjergji Kasneci Introduction to Information Retrieval WS

RETRIEVAL MODELS. Dr. Gjergji Kasneci Introduction to Information Retrieval WS RETRIEVAL MODELS Dr. Gjergji Kasneci Introduction to Information Retrieval WS 2012-13 1 Outline Intro Basics of probability and information theory Retrieval models Boolean model Vector space model Probabilistic

More information

Introduction to Logistic Regression

Introduction to Logistic Regression Introduction to Logistic Regression Guy Lebanon Binary Classification Binary classification is the most basic task in machine learning, and yet the most frequent. Binary classifiers often serve as the

More information

Advanced Machine Learning & Perception

Advanced Machine Learning & Perception Advanced Machine Learning & Perception Instructor: Tony Jebara Topic 6 Standard Kernels Unusual Input Spaces for Kernels String Kernels Probabilistic Kernels Fisher Kernels Probability Product Kernels

More information

Lecture 22 Exploratory Text Analysis & Topic Models

Lecture 22 Exploratory Text Analysis & Topic Models Lecture 22 Exploratory Text Analysis & Topic Models Intro to NLP, CS585, Fall 2014 http://people.cs.umass.edu/~brenocon/inlp2014/ Brendan O Connor [Some slides borrowed from Michael Paul] 1 Text Corpus

More information

Content-based Recommendation

Content-based Recommendation Content-based Recommendation Suthee Chaidaroon June 13, 2016 Contents 1 Introduction 1 1.1 Matrix Factorization......................... 2 2 slda 2 2.1 Model................................. 3 3 flda 3

More information

STA 4273H: Statistical Machine Learning

STA 4273H: Statistical Machine Learning STA 4273H: Statistical Machine Learning Russ Salakhutdinov Department of Statistics! rsalakhu@utstat.toronto.edu! http://www.utstat.utoronto.ca/~rsalakhu/ Sidney Smith Hall, Room 6002 Lecture 3 Linear

More information

Midterm sample questions

Midterm sample questions Midterm sample questions CS 585, Brendan O Connor and David Belanger October 12, 2014 1 Topics on the midterm Language concepts Translation issues: word order, multiword translations Human evaluation Parts

More information

MIDTERM: CS 6375 INSTRUCTOR: VIBHAV GOGATE October,

MIDTERM: CS 6375 INSTRUCTOR: VIBHAV GOGATE October, MIDTERM: CS 6375 INSTRUCTOR: VIBHAV GOGATE October, 23 2013 The exam is closed book. You are allowed a one-page cheat sheet. Answer the questions in the spaces provided on the question sheets. If you run

More information

Chapter 6 Classification and Prediction (2)

Chapter 6 Classification and Prediction (2) Chapter 6 Classification and Prediction (2) Outline Classification and Prediction Decision Tree Naïve Bayes Classifier Support Vector Machines (SVM) K-nearest Neighbors Accuracy and Error Measures Feature

More information

Worked Examples for Nominal Intercoder Reliability. by Deen G. Freelon October 30,

Worked Examples for Nominal Intercoder Reliability. by Deen G. Freelon October 30, Worked Examples for Nominal Intercoder Reliability by Deen G. Freelon (deen@dfreelon.org) October 30, 2009 http://www.dfreelon.com/utils/recalfront/ This document is an excerpt from a paper currently under

More information

Chapter 7 Forecasting Demand

Chapter 7 Forecasting Demand Chapter 7 Forecasting Demand Aims of the Chapter After reading this chapter you should be able to do the following: discuss the role of forecasting in inventory management; review different approaches

More information

CS838-1 Advanced NLP: Hidden Markov Models

CS838-1 Advanced NLP: Hidden Markov Models CS838-1 Advanced NLP: Hidden Markov Models Xiaojin Zhu 2007 Send comments to jerryzhu@cs.wisc.edu 1 Part of Speech Tagging Tag each word in a sentence with its part-of-speech, e.g., The/AT representative/nn

More information

CS 361: Probability & Statistics

CS 361: Probability & Statistics March 14, 2018 CS 361: Probability & Statistics Inference The prior From Bayes rule, we know that we can express our function of interest as Likelihood Prior Posterior The right hand side contains the

More information

Step-Stress Models and Associated Inference

Step-Stress Models and Associated Inference Department of Mathematics & Statistics Indian Institute of Technology Kanpur August 19, 2014 Outline Accelerated Life Test 1 Accelerated Life Test 2 3 4 5 6 7 Outline Accelerated Life Test 1 Accelerated

More information

CHAPTER-17. Decision Tree Induction

CHAPTER-17. Decision Tree Induction CHAPTER-17 Decision Tree Induction 17.1 Introduction 17.2 Attribute selection measure 17.3 Tree Pruning 17.4 Extracting Classification Rules from Decision Trees 17.5 Bayesian Classification 17.6 Bayes

More information

Lecture 13: Structured Prediction

Lecture 13: Structured Prediction Lecture 13: Structured Prediction Kai-Wei Chang CS @ University of Virginia kw@kwchang.net Couse webpage: http://kwchang.net/teaching/nlp16 CS6501: NLP 1 Quiz 2 v Lectures 9-13 v Lecture 12: before page

More information

arxiv: v3 [cs.lg] 25 Aug 2017

arxiv: v3 [cs.lg] 25 Aug 2017 Achieving Budget-optimality with Adaptive Schemes in Crowdsourcing Ashish Khetan and Sewoong Oh arxiv:602.0348v3 [cs.lg] 25 Aug 207 Abstract Crowdsourcing platforms provide marketplaces where task requesters

More information

Maxent Models and Discriminative Estimation

Maxent Models and Discriminative Estimation Maxent Models and Discriminative Estimation Generative vs. Discriminative models (Reading: J+M Ch6) Introduction So far we ve looked at generative models Language models, Naive Bayes But there is now much

More information

arxiv: v2 [cs.lg] 17 Nov 2016

arxiv: v2 [cs.lg] 17 Nov 2016 Approximating Wisdom of Crowds using K-RBMs Abhay Gupta Microsoft India R&D Pvt. Ltd. abhgup@microsoft.com arxiv:1611.05340v2 [cs.lg] 17 Nov 2016 Abstract An important way to make large training sets is

More information

Language Modelling: Smoothing and Model Complexity. COMP-599 Sept 14, 2016

Language Modelling: Smoothing and Model Complexity. COMP-599 Sept 14, 2016 Language Modelling: Smoothing and Model Complexity COMP-599 Sept 14, 2016 Announcements A1 has been released Due on Wednesday, September 28th Start code for Question 4: Includes some of the package import

More information

The OntoNL Semantic Relatedness Measure for OWL Ontologies

The OntoNL Semantic Relatedness Measure for OWL Ontologies The OntoNL Semantic Relatedness Measure for OWL Ontologies Anastasia Karanastasi and Stavros hristodoulakis Laboratory of Distributed Multimedia Information Systems and Applications Technical University

More information