A Probabilistic Approach for Integrating Heterogeneous Knowledge Sources

Size: px
Start display at page:

Download "A Probabilistic Approach for Integrating Heterogeneous Knowledge Sources"

Transcription

1 A Probabilistic Approach for Integrating Heterogeneous Knowledge Sources Arnab Dutta, Christian Meilicke, Simone Ponzetto

2 Which Superman exactly? Brandon Routh was an actor in Superman

3 Uncertainty in the Web Underspecified terms Which superman? Absence of clear semantics Is superman always the 1983 movie in all triples it occurs?

4 Information Extraction (IE) Projects Closed (e.g. DBpedia, Yago) maintain a well-defined schema URI used to uniquely identify an entity x Restricted to Wikipedia mostly Open (e.g. Nell, ReVerb) work on large sets of unstructured data x necessarily do not maintain a schema domain independent scale easily to web corpus

5 NELL can someone semantify me please? I can offer new facts that I have learnt from the entire Web I can help you. You can use my ontology. But enrich me, I am just restricted to Wikipedia DBpedia

6 Approach Baseline Most frequent sense based approach Probabilistic exploits additional information to make intelligent choices

7 Approach: Baseline Exploit Wikipedia Corpus (used Wikiprep) Anchors link to pages, often multiple ones Anchors analogous to NELL terms

8 Approach: Baseline Exploit Wikipedia Corpus (used Wikiprep) Anchors link to pages, often multiple ones Anchors analogous to NELL terms

9 Approach: Baseline Exploit Wikipedia Corpus (used Wikiprep) Anchors link to pages, often multiple ones Anchors analogous to NELL terms

10 Approach: Baseline (contd.) Compute probability from the anchor link counts to pages Get top-k candidates from the distribution Page Title Link count Superman 4133 Superman (film) 521 Superman (comic book) 491

11 Baseline drawbacks Not exploiting the context Most frequent sense is not always the intended reference

12 Approach: Probabilistic define a set of matching hypotheses exploit the ontology of DBpedia penalize the wrong hypotheses if it leads to any inconsistency

13 Big Picture Brandon routh actorstarredinmovie superman

14 Big Picture Brandon routh actorstarredinmovie superman db:brandon_routh db:superman Db:Superman_(comic_book) db:superman_(film)

15 Big Picture Brandon routh actorstarredinmovie superman db:brandon_routh db:superman Db:Superman_(comic_book) db:superman_(film)

16 Big Picture Brandon routh actorstarredinmovie superman db:brandon_routh db:superman Db:Superman_(comic_book) db:superman_(film)

17 Methodology I. Probabilistic type generation range(actorstarredinmovie, Person, w 1 ) range(actorstarredinmovie, Artist, w 2 ) range(actorstarredinmovie, Athlete, w 3 ) II. Formulating as an inference task in a Markov Network A state which is most likely to hold

18 I. Probabilistic Type Generation

19 I. Probabilistic Type Generation Scan through all the NELL triples <, actorstarredinmovie, > <brandon routh, actorstarredinmovie, superman>

20 I. Probabilistic Type Generation Scan through all the NELL triples <, actorstarredinmovie, > <brandon routh, actorstarredinmovie, superman> Get the top-1 references of the instances db:brandon_routh db:superman

21 I. Probabilistic Type Generation Scan through all the NELL triples <, actorstarredinmovie, > <brandon routh, actorstarredinmovie, superman> Get the top-1 references of the instances db:brandon_routh db:superman Get the direct types for them dbo:person dbo:fictionalcharacter

22 Tree Generation Work (100) Direct type counts Cartoon (20) Film (40) WrittenWork (80) Book (30) Play (20) Comics (10) Selection Preference: Film > Book > Play = Cartoon >..

23 Node Scoring Up-score V u (n)= V o (n) + α * Σ V u (n) child of n Down-score V d (n)= V d (parent(n)) + (1-α) V u (n)

24 Tree Generation (α-tree) Person (100) V d Cartoon (20) Film (40) WrittenWork (80) Book (30) Play (20) Comics (10)

25 Tree Generation (α-tree) Person (100, (100) 185) V u V d Cartoon (20, (20) 20) Film (40, (40) 40) WrittenWork (80, (80) 110) Book (30, (30) 30) Play (20, (20) 20) Comics (10, (10) 10)

26 Tree Generation (α-tree) Person Person (100, (100) 185) (100, 185, 185) V u V d Cartoon Cartoon (20, (20) 20) (20, 20, 195) Film Film (40, (40) 40) (40, 40, 205) WrittenWork WrittenWork (80, (80) 110) (80, 110, 240) Book Book (30, (30) 30) (30, 30, 255) Play Play (20, (20) 20) (20, 20, 250) Comics Comics (10, (10) 10) (10, 10, 245) Selection Preference: Book > Play > Comics >..

27 II. Formulate as a Markov Logic Network

28 II. Formulate as a Markov Logic Network Add weighted/un-weighted first order rules Un-weighted sameas(a,b) Λ sameas(b, c) => sameas(a, c) x sameas(x, y) <= 1 weighted w: nelltriple(s, prop, o) Λ sameasconf(o, db:instance, conf) Λ istype(db:instance, db:class) => sameas(o, db:instance)

29 II. Formulate as a Markov Logic Network Add weighted/un-weighted first order rules Un-weighted sameas(a,b) Λ sameas(b, c) => sameas(a, c) x sameas(x, y) <= 1 weighted w: nelltriple(s, prop, o) Λ sameasconf(o, db:instance, conf) Λ istype(db:instance, db:class) => sameas(o, db:instance) Add soft truths as observed (evidences) sameasconf(superman, db:superman, w1) sameasconf (superman, db:superman_(film), w2) sameasconf (superman, db:superman_(comic_book), w3) and many more. Add hard truths istype(db:superman, ComicsCharacter) istype(db:superman_(film), Film)

30 II. Formulate as a Markov Logic Network Add weighted/un-weighted first order rules Un-weighted sameas(a,b) Λ sameas(b, c) => sameas(a, c) x sameas(x, y) <= 1 weighted w: nelltriple(s, prop, o) Λ sameasconf(o, db:instance, conf) Λ istype(db:instance, db:class) => sameas(o, db:instance) Add soft truths as observed (evidences) sameasconf(superman, db:superman, w1) sameasconf (superman, db:superman_(film), w2) sameasconf (superman, db:superman_(comic_book), w3) and many more. Add hard truths istype(db:superman, ComicsCharacter) istype(db:superman_(film), Film) Ask for the most likely state of the network MAP query (RockIt system used as the engine)

31 Grounding the Markov Network

32 Grounding the Markov Network istype(dbinst,c) sameasconf(o, dbinst, conf) nelltriple(s, p, o)

33 Grounding the Markov Network istype(dbinst,c) sameasconf(o, dbinst, conf) nelltriple(s, p, o) W 11 : istype(db:superman_(film), Film) Λ nelltriple(s, p, superman) Λ sameasconf(superman, db:superman_(film), w2) => sameas(superman, db:superman_(film))

34 Grounding the Markov Network istype(dbinst,c) sameasconf(o, dbinst, conf) nelltriple(s, p, o) W 11 : istype(db:superman_(film), Film) Λ nelltriple(s, p, superman) Λ sameasconf(superman, db:superman_(film), w2) => sameas(superman, db:superman_(film)) W 12 : istype(db:superman_(film), WrittenWork) Λ nelltriple(s, p, superman) ) Λ sameasconf(superman, db:superman_(film), w2) => sameas (superman, db:superman_(film))

35 Grounding the Markov Network istype(dbinst,c) sameasconf(o, dbinst, conf) nelltriple(s, p, o) W 11 : istype(db:superman_(film), Film) Λ nelltriple(s, p, superman) Λ sameasconf(superman, db:superman_(film), w2) => sameas(superman, db:superman_(film)) W 12 : istype(db:superman_(film), WrittenWork) Λ nelltriple(s, p, superman) ) Λ sameasconf(superman, db:superman_(film), w2) => sameas (superman, db:superman_(film))... and many many more such instantiations

36 Grounding the Markov Network istype(dbinst,c) sameasconf(o, dbinst, conf) nelltriple(s, p, o) W 11 : istype(db:superman_(film), Film) Λ nelltriple(s, p, superman) Λ sameasconf(superman, db:superman_(film), w2) => sameas(superman, db:superman_(film)) W 12 : istype(db:superman_(film), WrittenWork) Λ nelltriple(s, p, superman) ) Λ sameasconf(superman, db:superman_(film), w2) => sameas (superman, db:superman_(film))... and many many more such instantiations

37 Can we still do better? If the output is a refinement, can t we use it again to generate type weights again?

38 Bootstrapping Take all the sameas hypotheses Includes the correct and incorrect mappings Use the top-1 mapping to create type weights Assumption is the most frequent mapping is correct Add the sameas and the type weights to compete Result is a better mapping set (refined than top-1) Use it again to get the type weights.

39 Bootstrapping in action

40 Bootstrapping in action All same as hypotheses sameas(superman, db:superman, 1.9) sameas(superman, db:superman_(film), 1.3)

41 Bootstrapping in action All same as hypotheses sameas(superman, db:superman, 1.9) sameas(superman, db:superman_(film), 1.3) Type weights from top-1 only

42 Bootstrapping in action All same as hypotheses sameas(superman, db:superman, 1.9) sameas(superman, db:superman_(film), 1.3) Type Weights Type weights from top-1 only 2.3: Work -1.1: WrittenWork -2.2: Film -3.9: FictionalCharacter

43 Bootstrapping in action All same as hypotheses sameas(superman, db:superman, 1.9) sameas(superman, db:superman_(film), 1.3) Type Weights Type weights from top-1 only 2.3: Work -1.1: WrittenWork -2.2: Film -3.9: FictionalCharacter Refined mappings

44 Bootstrapping in action All same as hypotheses sameas(superman, db:superman, 1.9) sameas(superman, db:superman_(film), 1.3) Type Weights Type weights from top-1 only 2.3: Work -1.1: WrittenWork -2.2: Film -3.9: FictionalCharacter No suitable mappings Refined mappings

45 Bootstrapping in action All same as hypotheses sameas(superman, db:superman, 1.9) sameas(superman, db:superman_(film), 1.3) Feed again Type weights from top-1 only Type Weights 2.3: Work -1.1: WrittenWork -2.2: Film -3.9: FictionalCharacter No suitable mappings Refined mappings

46 Bootstrapping in action All same as hypotheses sameas(superman, db:superman, 1.9) sameas(superman, db:superman_(film), 1.3) Feed again Type weights from top-1 only Type Weights 2.3: Work -1.1: WrittenWork -2.2: Film -3.9: FictionalCharacter No suitable mappings Refined mappings

47 Bootstrapping in action All same as hypotheses sameas(superman, db:superman, 1.9) sameas(superman, db:superman_(film), 1.3) Feed again Type weights from refined maps only Type Weights 2.3: Work -1.1: WrittenWork -2.2: Film -3.9: FictionalCharacter Refined mappings No suitable mappings

48 Bootstrapping in action All same as hypotheses sameas(superman, db:superman, 1.9) sameas(superman, db:superman_(film), 1.3) Feed again Type weights from refined maps only Type Weights 2.3: Type WorkWeights recomputed -1.1: WrittenWork -2.2: 3.5: Film Film 1.0: Work -3.9: FictionalCharacter -2.3: Politician Refined mappings No suitable mappings

49 Bootstrapping in action All same as hypotheses sameas(superman, db:superman, 1.9) sameas(superman, db:superman_(film), 1.3) Feed again Type weights from refined maps only Type Weights 2.3: Type WorkWeights recomputed -1.1: WrittenWork -2.2: 3.5: Film Film 1.0: Work -3.9: FictionalCharacter -2.3: Politician Refined mappings No suitable mappings sameas(superman, db:superman_(film))

50 Experiments Nell data set consisting of 100 triples for 12 randomly chosen properties Manually create a gold standard Learn the parameter alpha (α)

51 Learning α Divide α in steps of between [0, 1] Large number of samples for 6 properties, D train Get the maximum α on the D train Apply the α on the rest of the 6 properties, D test 35% case α = % case α = % cases α = 0.625

52 Variation of scores with α

53 Comparative Values (α = 0.5)

54 Comparative Values (α = 0.5)

55 Limitations Unable to disambiguate instances with similar types. e.g. Mel Gibson and William Gibson, both of type dbo:person Missing DBpedia type information

56 Looking ahead Mapping of properties as well Potential scope of generating new knowledge Apply on ReVerb data set as well

57 Thank You

Information Extraction from Text

Information Extraction from Text Information Extraction from Text Jing Jiang Chapter 2 from Mining Text Data (2012) Presented by Andrew Landgraf, September 13, 2013 1 What is Information Extraction? Goal is to discover structured information

More information

CIKM 18, October 22-26, 2018, Torino, Italy

CIKM 18, October 22-26, 2018, Torino, Italy 903 Session 6B: Knowledge Modelling 1.1 Entity Typing by Structured Features Existing systems [16, 19, 27] cast entity typing as a single-instance multi-label classification problem. For each entity e,

More information

Topic Models and Applications to Short Documents

Topic Models and Applications to Short Documents Topic Models and Applications to Short Documents Dieu-Thu Le Email: dieuthu.le@unitn.it Trento University April 6, 2011 1 / 43 Outline Introduction Latent Dirichlet Allocation Gibbs Sampling Short Text

More information

UnSAID: Uncertainty and Structure in the Access to Intensional Data

UnSAID: Uncertainty and Structure in the Access to Intensional Data UnSAID: Uncertainty and Structure in the Access to Intensional Data Pierre Senellart 3 July 214, Univ. Rennes 1 Uncertain data is everywhere Numerous sources of uncertain data: Measurement errors Data

More information

Data-Driven Logical Reasoning

Data-Driven Logical Reasoning Data-Driven Logical Reasoning Claudia d Amato Volha Bryl, Luciano Serafini November 11, 2012 8 th International Workshop on Uncertainty Reasoning for the Semantic Web 11 th ISWC, Boston (MA), USA. Heterogeneous

More information

DBpedia Ontology Enrichment for Inconsistency Detection and Statistical Schema Induction

DBpedia Ontology Enrichment for Inconsistency Detection and Statistical Schema Induction DBpedia Ontology Enrichment for Inconsistency Detection and Statistical Schema Induction Presentation By Tung Do 1. Statistical Schema Induction 2. DBpedia Ontology Enrichment for Inconsistency Detection

More information

PROBABILISTIC KNOWLEDGE GRAPH CONSTRUCTION: COMPOSITIONAL AND INCREMENTAL APPROACHES. Dongwoo Kim with Lexing Xie and Cheng Soon Ong CIKM 2016

PROBABILISTIC KNOWLEDGE GRAPH CONSTRUCTION: COMPOSITIONAL AND INCREMENTAL APPROACHES. Dongwoo Kim with Lexing Xie and Cheng Soon Ong CIKM 2016 PROBABILISTIC KNOWLEDGE GRAPH CONSTRUCTION: COMPOSITIONAL AND INCREMENTAL APPROACHES Dongwoo Kim with Lexing Xie and Cheng Soon Ong CIKM 2016 1 WHAT IS KNOWLEDGE GRAPH (KG)? KG widely used in various tasks

More information

Prof. Dr. Ralf Möller Dr. Özgür L. Özçep Universität zu Lübeck Institut für Informationssysteme. Tanya Braun (Exercises)

Prof. Dr. Ralf Möller Dr. Özgür L. Özçep Universität zu Lübeck Institut für Informationssysteme. Tanya Braun (Exercises) Prof. Dr. Ralf Möller Dr. Özgür L. Özçep Universität zu Lübeck Institut für Informationssysteme Tanya Braun (Exercises) Slides taken from the presentation (subset only) Learning Statistical Models From

More information

Mihai Boicu, Gheorghe Tecuci, Dorin Marcu Learning Agent Center, George Mason University

Mihai Boicu, Gheorghe Tecuci, Dorin Marcu Learning Agent Center, George Mason University Mihai Boicu, Gheorghe Tecuci, Dorin Marcu Learning Agent Center, George Mason University The Seventh International Conference on Semantic Technologies for Intelligence, Defense, and Security STIDS Fairfax,

More information

Integrating Order Information and Event Relation for Script Event Prediction

Integrating Order Information and Event Relation for Script Event Prediction Integrating Order Information and Event Relation for Script Event Prediction Zhongqing Wang 1,2, Yue Zhang 2 and Ching-Yun Chang 2 1 Soochow University, China 2 Singapore University of Technology and Design

More information

Probabilistic Logic CNF for Reasoning

Probabilistic Logic CNF for Reasoning Probabilistic Logic CNF for Reasoning Song- Chun Zhu, Sinisa Todorovic, and Ales Leonardis At CVPR, Providence, Rhode Island June 16, 2012 Goal input tracking parsing reasoning activity recognition in

More information

Propositions. c D. Poole and A. Mackworth 2010 Artificial Intelligence, Lecture 5.1, Page 1

Propositions. c D. Poole and A. Mackworth 2010 Artificial Intelligence, Lecture 5.1, Page 1 Propositions An interpretation is an assignment of values to all variables. A model is an interpretation that satisfies the constraints. Often we don t want to just find a model, but want to know what

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

Probability, Entropy, and Inference / More About Inference

Probability, Entropy, and Inference / More About Inference Probability, Entropy, and Inference / More About Inference Mário S. Alvim (msalvim@dcc.ufmg.br) Information Theory DCC-UFMG (2018/02) Mário S. Alvim (msalvim@dcc.ufmg.br) Probability, Entropy, and Inference

More information

Semantic Web SPARQL. Gerd Gröner, Matthias Thimm. July 16,

Semantic Web SPARQL. Gerd Gröner, Matthias Thimm. July 16, Semantic Web SPARQL Gerd Gröner, Matthias Thimm {groener,thimm}@uni-koblenz.de Institute for Web Science and Technologies (WeST) University of Koblenz-Landau July 16, 2013 Gerd Gröner, Matthias Thimm Semantic

More information

Knowledge Expansion over Probabilistic Knowledge Bases. Yang Chen, Daisy Zhe Wang

Knowledge Expansion over Probabilistic Knowledge Bases. Yang Chen, Daisy Zhe Wang Knowledge Expansion over Probabilistic Knowledge Bases Yang Chen, Daisy Zhe Wang {yang,daisyw}cise.ufl.edu Computer and Information Science and Engineering University of Florida ProbKB SIGMOD, Snowbird,

More information

Stat 315c: Introduction

Stat 315c: Introduction Stat 315c: Introduction Art B. Owen Stanford Statistics Art B. Owen (Stanford Statistics) Stat 315c: Introduction 1 / 14 Stat 315c Analysis of Transposable Data Usual Statistics Setup there s Y (we ll

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

PROBABILISTIC LOGIC. J. Webster (ed.), Wiley Encyclopedia of Electrical and Electronics Engineering Copyright c 1999 John Wiley & Sons, Inc.

PROBABILISTIC LOGIC. J. Webster (ed.), Wiley Encyclopedia of Electrical and Electronics Engineering Copyright c 1999 John Wiley & Sons, Inc. J. Webster (ed.), Wiley Encyclopedia of Electrical and Electronics Engineering Copyright c 1999 John Wiley & Sons, Inc. PROBABILISTIC LOGIC A deductive argument is a claim of the form: If P 1, P 2,...,andP

More information

CS 6375 Machine Learning

CS 6375 Machine Learning CS 6375 Machine Learning Decision Trees Instructor: Yang Liu 1 Supervised Classifier X 1 X 2. X M Ref class label 2 1 Three variables: Attribute 1: Hair = {blond, dark} Attribute 2: Height = {tall, short}

More information

A Convolutional Neural Network-based

A Convolutional Neural Network-based A Convolutional Neural Network-based Model for Knowledge Base Completion Dat Quoc Nguyen Joint work with: Dai Quoc Nguyen, Tu Dinh Nguyen and Dinh Phung April 16, 2018 Introduction Word vectors learned

More information

Artificial Intelligence Bayes Nets: Independence

Artificial Intelligence Bayes Nets: Independence Artificial Intelligence Bayes Nets: Independence Instructors: David Suter and Qince Li Course Delivered @ Harbin Institute of Technology [Many slides adapted from those created by Dan Klein and Pieter

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

Machine Learning 4771

Machine Learning 4771 Machine Learning 4771 Instructor: Tony Jebara Topic 16 Undirected Graphs Undirected Separation Inferring Marginals & Conditionals Moralization Junction Trees Triangulation Undirected Graphs Separation

More information

Topic-based Selectivity Estimation for Hybrid Queries over RDF Graphs Technical Report

Topic-based Selectivity Estimation for Hybrid Queries over RDF Graphs Technical Report Topic-based Selectivity Estimation for Hybrid Queries over RDF Graphs Technical Report Andreas Wagner AIFB, KIT Karlsruhe, Germany a.wagner@kit.edu Veli Bicer IBM Research, Smarter Cities Technology Centre

More information

Provenance-Aware Entity Resolution Leveraging Provenance to Improve Quality

Provenance-Aware Entity Resolution Leveraging Provenance to Improve Quality ProvenanceAware Entity Resolution Leveraging Provenance to Improve Quality Qing Wang, KlausDieter Schewe 2 and Woods Wang Research School of Computer Science, Australian National University, Australia

More information

Lecture 4: Probability Propagation in Singly Connected Bayesian Networks

Lecture 4: Probability Propagation in Singly Connected Bayesian Networks Lecture 4: Probability Propagation in Singly Connected Bayesian Networks Singly Connected Networks So far we have looked at several cases of probability propagation in networks. We now want to formalise

More information

Bayesian Identity Clustering

Bayesian Identity Clustering Bayesian Identity Clustering Simon JD Prince Department of Computer Science University College London James Elder Centre for Vision Research York University http://pvlcsuclacuk sprince@csuclacuk The problem

More information

Course 395: Machine Learning

Course 395: Machine Learning Course 395: Machine Learning Lecturers: Maja Pantic (maja@doc.ic.ac.uk) Stavros Petridis (sp104@doc.ic.ac.uk) Goal (Lectures): To present basic theoretical concepts and key algorithms that form the core

More information

Strategies for Geographical Scoping and Improving a Gazetteer

Strategies for Geographical Scoping and Improving a Gazetteer Strategies for Geographical Scoping and Improving a Gazetteer ABSTRACT Sanket Kumar Singh University of Alberta sanketku@ualberta.ca Many applications that use geographical databases (a.k.a. gazetteers)

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

Logic Programming Techniques for Reasoning with Probabilistic Ontologies

Logic Programming Techniques for Reasoning with Probabilistic Ontologies Logic Programming Techniques for Reasoning with Probabilistic Ontologies Riccardo Zese, Elena Bellodi, Evelina Lamma and Fabrizio Riguzzi University of Ferrara, Italy riccardo.zese@unife.it Zese, Bellodi,

More information

Propositional Logic: Models and Proofs

Propositional Logic: Models and Proofs Propositional Logic: Models and Proofs C. R. Ramakrishnan CSE 505 1 Syntax 2 Model Theory 3 Proof Theory and Resolution Compiled at 11:51 on 2016/11/02 Computing with Logic Propositional Logic CSE 505

More information

CS 5522: Artificial Intelligence II

CS 5522: Artificial Intelligence II CS 5522: Artificial Intelligence II Bayes Nets: Independence Instructor: Alan Ritter Ohio State University [These slides were adapted from CS188 Intro to AI at UC Berkeley. All materials available at http://ai.berkeley.edu.]

More information

Ch.6 Uncertain Knowledge. Logic and Uncertainty. Representation. One problem with logical approaches: Department of Computer Science

Ch.6 Uncertain Knowledge. Logic and Uncertainty. Representation. One problem with logical approaches: Department of Computer Science Ch.6 Uncertain Knowledge Representation Hantao Zhang http://www.cs.uiowa.edu/ hzhang/c145 The University of Iowa Department of Computer Science Artificial Intelligence p.1/39 Logic and Uncertainty One

More information

ECE 6504: Advanced Topics in Machine Learning Probabilistic Graphical Models and Large-Scale Learning

ECE 6504: Advanced Topics in Machine Learning Probabilistic Graphical Models and Large-Scale Learning ECE 6504: Advanced Topics in Machine Learning Probabilistic Graphical Models and Large-Scale Learning Topics Summary of Class Advanced Topics Dhruv Batra Virginia Tech HW1 Grades Mean: 28.5/38 ~= 74.9%

More information

Semantics and Inference for Probabilistic Ontologies

Semantics and Inference for Probabilistic Ontologies Semantics and Inference for Probabilistic Ontologies Fabrizio Riguzzi, Elena Bellodi, Evelina Lamma, and Riccardo Zese ENDIF University of Ferrara, Italy, email: {fabrizio.riguzzi, elena.bellodi, evelina.lamma}@unife.it,

More information

Parts 3-6 are EXAMPLES for cse634

Parts 3-6 are EXAMPLES for cse634 1 Parts 3-6 are EXAMPLES for cse634 FINAL TEST CSE 352 ARTIFICIAL INTELLIGENCE Fall 2008 There are 6 pages in this exam. Please make sure you have all of them INTRODUCTION Philosophical AI Questions Q1.

More information

PROBABILISTIC REASONING SYSTEMS

PROBABILISTIC REASONING SYSTEMS PROBABILISTIC REASONING SYSTEMS In which we explain how to build reasoning systems that use network models to reason with uncertainty according to the laws of probability theory. Outline Knowledge in uncertain

More information

Bayesian Networks: Construction, Inference, Learning and Causal Interpretation. Volker Tresp Summer 2016

Bayesian Networks: Construction, Inference, Learning and Causal Interpretation. Volker Tresp Summer 2016 Bayesian Networks: Construction, Inference, Learning and Causal Interpretation Volker Tresp Summer 2016 1 Introduction So far we were mostly concerned with supervised learning: we predicted one or several

More information

Understanding Interlinked Data

Understanding Interlinked Data Understanding Interlinked Data Visualising, Exploring, and Analysing Ontologies Olaf Noppens and Thorsten Liebig (Ulm University, Germany olaf.noppens@uni-ulm.de, thorsten.liebig@uni-ulm.de) Abstract Companies

More information

Bayesian Networks: Construction, Inference, Learning and Causal Interpretation. Volker Tresp Summer 2014

Bayesian Networks: Construction, Inference, Learning and Causal Interpretation. Volker Tresp Summer 2014 Bayesian Networks: Construction, Inference, Learning and Causal Interpretation Volker Tresp Summer 2014 1 Introduction So far we were mostly concerned with supervised learning: we predicted one or several

More information

A Review of Relational Machine Learning for Knowledge Graphs

A Review of Relational Machine Learning for Knowledge Graphs CBMM Memo No. 28 March 23, 2015 A Review of Relational Machine Learning for Knowledge Graphs From Multi-Relational Link Prediction to Automated Knowledge Graph Construction by Maximilian Nickel, Kevin

More information

Respecting Markov Equivalence in Computing Posterior Probabilities of Causal Graphical Features

Respecting Markov Equivalence in Computing Posterior Probabilities of Causal Graphical Features Proceedings of the Twenty-Fourth AAAI Conference on Artificial Intelligence (AAAI-10) Respecting Markov Equivalence in Computing Posterior Probabilities of Causal Graphical Features Eun Yong Kang Department

More information

Modal Logics. Most applications of modal logic require a refined version of basic modal logic.

Modal Logics. Most applications of modal logic require a refined version of basic modal logic. Modal Logics Most applications of modal logic require a refined version of basic modal logic. Definition. A set L of formulas of basic modal logic is called a (normal) modal logic if the following closure

More information

BITS F464: MACHINE LEARNING

BITS F464: MACHINE LEARNING BITS F464: MACHINE LEARNING Lecture-09: Concept Learning Dr. Kamlesh Tiwari Assistant Professor Department of Computer Science and Information Systems Engineering, BITS Pilani, Rajasthan-333031 INDIA Jan

More information

Conditional Inclusion Dependencies

Conditional Inclusion Dependencies Conditional Inclusion Dependencies Jana Bauckmann Guest lecture in Data Profiling and Data Cleansing Prof. Dr. Felix Naumann Outline 2 Defining Conditional Inclusion Dependencies (CINDs) Fields of Application

More information

26 : Spectral GMs. Lecturer: Eric P. Xing Scribes: Guillermo A Cidre, Abelino Jimenez G.

26 : Spectral GMs. Lecturer: Eric P. Xing Scribes: Guillermo A Cidre, Abelino Jimenez G. 10-708: Probabilistic Graphical Models, Spring 2015 26 : Spectral GMs Lecturer: Eric P. Xing Scribes: Guillermo A Cidre, Abelino Jimenez G. 1 Introduction A common task in machine learning is to work with

More information

INFO 4300 / CS4300 Information Retrieval. slides adapted from Hinrich Schütze s, linked from

INFO 4300 / CS4300 Information Retrieval. slides adapted from Hinrich Schütze s, linked from INFO 4300 / CS4300 Information Retrieval slides adapted from Hinrich Schütze s, linked from http://informationretrieval.org/ IR 8: Evaluation & SVD Paul Ginsparg Cornell University, Ithaca, NY 20 Sep 2011

More information

JOINT PROBABILISTIC INFERENCE OF CAUSAL STRUCTURE

JOINT PROBABILISTIC INFERENCE OF CAUSAL STRUCTURE JOINT PROBABILISTIC INFERENCE OF CAUSAL STRUCTURE Dhanya Sridhar Lise Getoor U.C. Santa Cruz KDD Workshop on Causal Discovery August 14 th, 2016 1 Outline Motivation Problem Formulation Our Approach Preliminary

More information

Information Retrieval and Organisation

Information Retrieval and Organisation Information Retrieval and Organisation Chapter 13 Text Classification and Naïve Bayes Dell Zhang Birkbeck, University of London Motivation Relevance Feedback revisited The user marks a number of documents

More information

Discrete Structures for Computer Science

Discrete Structures for Computer Science Discrete Structures for Computer Science William Garrison bill@cs.pitt.edu 6311 Sennott Square Lecture #6: Rules of Inference Based on materials developed by Dr. Adam Lee Today s topics n Rules of inference

More information

day month year documentname/initials 1

day month year documentname/initials 1 ECE471-571 Pattern Recognition Lecture 13 Decision Tree Hairong Qi, Gonzalez Family Professor Electrical Engineering and Computer Science University of Tennessee, Knoxville http://www.eecs.utk.edu/faculty/qi

More information

Language Models. CS6200: Information Retrieval. Slides by: Jesse Anderton

Language Models. CS6200: Information Retrieval. Slides by: Jesse Anderton Language Models CS6200: Information Retrieval Slides by: Jesse Anderton What s wrong with VSMs? Vector Space Models work reasonably well, but have a few problems: They are based on bag-of-words, so they

More information

Fielded Sequential Dependence Model for Ad-Hoc Entity Retrieval in the Web of Data

Fielded Sequential Dependence Model for Ad-Hoc Entity Retrieval in the Web of Data Fielded Sequential Dependence Model for Ad-Hoc Entity Retrieval in the Web of Data Nikita Zhiltsov 1,2 Alexander Kotov 3 Fedor Nikolaev 3 1 Kazan Federal University 2 Textocat 3 Textual Data Analytics

More information

A Regularized Recommendation Algorithm with Probabilistic Sentiment-Ratings

A Regularized Recommendation Algorithm with Probabilistic Sentiment-Ratings A Regularized Recommendation Algorithm with Probabilistic Sentiment-Ratings Filipa Peleja, Pedro Dias and João Magalhães Department of Computer Science Faculdade de Ciências e Tecnologia Universidade Nova

More information

Challenge Paper: Marginal Probabilities for Instances and Classes (Poster Presentation SRL Workshop)

Challenge Paper: Marginal Probabilities for Instances and Classes (Poster Presentation SRL Workshop) (Poster Presentation SRL Workshop) Oliver Schulte School of Computing Science, Simon Fraser University, Vancouver-Burnaby, Canada Abstract In classic AI research on combining logic and probability, Halpern

More information

Feedback Loop between High Level Semantics and Low Level Vision

Feedback Loop between High Level Semantics and Low Level Vision Feedback Loop between High Level Semantics and Low Level Vision Varun K. Nagaraja Vlad I. Morariu Larry S. Davis {varun,morariu,lsd}@umiacs.umd.edu University of Maryland, College Park, MD, USA. Abstract.

More information

Combining Sum-Product Network and Noisy-Or Model for Ontology Matching

Combining Sum-Product Network and Noisy-Or Model for Ontology Matching Combining Sum-Product Network and Noisy-Or Model for Ontology Matching Weizhuo Li Institute of Mathematics,Academy of Mathematics and Systems Science, Chinese Academy of Sciences, Beijing, P. R. China

More information

EE562 ARTIFICIAL INTELLIGENCE FOR ENGINEERS

EE562 ARTIFICIAL INTELLIGENCE FOR ENGINEERS EE562 ARTIFICIAL INTELLIGENCE FOR ENGINEERS Lecture 16, 6/1/2005 University of Washington, Department of Electrical Engineering Spring 2005 Instructor: Professor Jeff A. Bilmes Uncertainty & Bayesian Networks

More information

Sum-Product Networks: A New Deep Architecture

Sum-Product Networks: A New Deep Architecture Sum-Product Networks: A New Deep Architecture Pedro Domingos Dept. Computer Science & Eng. University of Washington Joint work with Hoifung Poon 1 Graphical Models: Challenges Bayesian Network Markov Network

More information

Jeffrey D. Ullman Stanford University

Jeffrey D. Ullman Stanford University Jeffrey D. Ullman Stanford University 3 We are given a set of training examples, consisting of input-output pairs (x,y), where: 1. x is an item of the type we want to evaluate. 2. y is the value of some

More information

Computational Genomics. Systems biology. Putting it together: Data integration using graphical models

Computational Genomics. Systems biology. Putting it together: Data integration using graphical models 02-710 Computational Genomics Systems biology Putting it together: Data integration using graphical models High throughput data So far in this class we discussed several different types of high throughput

More information

Selectivity Estimation for Hybrid Queries over Text-Rich Data Graphs

Selectivity Estimation for Hybrid Queries over Text-Rich Data Graphs Selectivity Estimation for Hybrid Queries over Text-Rich Data Graphs Andreas Wagner AIFB, KIT Karlsruhe, Germany a.wagner@kit.edu Veli Bicer IBM Research, Smarter Cities Technology Centre Dublin, Ireland

More information

[read Chapter 2] [suggested exercises 2.2, 2.3, 2.4, 2.6] General-to-specific ordering over hypotheses

[read Chapter 2] [suggested exercises 2.2, 2.3, 2.4, 2.6] General-to-specific ordering over hypotheses 1 CONCEPT LEARNING AND THE GENERAL-TO-SPECIFIC ORDERING [read Chapter 2] [suggested exercises 2.2, 2.3, 2.4, 2.6] Learning from examples General-to-specific ordering over hypotheses Version spaces and

More information

One-Way ANOVA. Some examples of when ANOVA would be appropriate include:

One-Way ANOVA. Some examples of when ANOVA would be appropriate include: One-Way ANOVA 1. Purpose Analysis of variance (ANOVA) is used when one wishes to determine whether two or more groups (e.g., classes A, B, and C) differ on some outcome of interest (e.g., an achievement

More information

A Tutorial on Learning with Bayesian Networks

A Tutorial on Learning with Bayesian Networks A utorial on Learning with Bayesian Networks David Heckerman Presented by: Krishna V Chengavalli April 21 2003 Outline Introduction Different Approaches Bayesian Networks Learning Probabilities and Structure

More information

Representing and Querying Correlated Tuples in Probabilistic Databases

Representing and Querying Correlated Tuples in Probabilistic Databases Representing and Querying Correlated Tuples in Probabilistic Databases Prithviraj Sen Amol Deshpande Department of Computer Science University of Maryland, College Park. International Conference on Data

More information

CS246 Final Exam, Winter 2011

CS246 Final Exam, Winter 2011 CS246 Final Exam, Winter 2011 1. Your name and student ID. Name:... Student ID:... 2. I agree to comply with Stanford Honor Code. Signature:... 3. There should be 17 numbered pages in this exam (including

More information

Abstract. Acknowledgements. 1 Introduction The focus of this dissertation Reinforcement Learning... 4

Abstract. Acknowledgements. 1 Introduction The focus of this dissertation Reinforcement Learning... 4 Contents iii Abstract i Acknowledgements ii 1 Introduction 1 1.1 The focus of this dissertation....................... 2 1.2 Reinforcement Learning.......................... 4 1.2.1 A qualitative introduction.....................

More information

Ordering, Indexing, and Searching Semantic Data: A Terminology Aware Index Structure

Ordering, Indexing, and Searching Semantic Data: A Terminology Aware Index Structure Ordering, Indexing, and Searching Semantic Data: A Terminology Aware Index Structure by Jeffrey Pound A thesis presented to the University of Waterloo in fulfillment of the thesis requirement for the degree

More information

6.867 Machine learning, lecture 23 (Jaakkola)

6.867 Machine learning, lecture 23 (Jaakkola) Lecture topics: Markov Random Fields Probabilistic inference Markov Random Fields We will briefly go over undirected graphical models or Markov Random Fields (MRFs) as they will be needed in the context

More information

Citation for published version (APA): Andogah, G. (2010). Geographically constrained information retrieval Groningen: s.n.

Citation for published version (APA): Andogah, G. (2010). Geographically constrained information retrieval Groningen: s.n. University of Groningen Geographically constrained information retrieval Andogah, Geoffrey IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from

More information

Programming Assignment 4: Image Completion using Mixture of Bernoullis

Programming Assignment 4: Image Completion using Mixture of Bernoullis Programming Assignment 4: Image Completion using Mixture of Bernoullis Deadline: Tuesday, April 4, at 11:59pm TA: Renie Liao (csc321ta@cs.toronto.edu) Submission: You must submit two files through MarkUs

More information

Classification Based on Logical Concept Analysis

Classification Based on Logical Concept Analysis Classification Based on Logical Concept Analysis Yan Zhao and Yiyu Yao Department of Computer Science, University of Regina, Regina, Saskatchewan, Canada S4S 0A2 E-mail: {yanzhao, yyao}@cs.uregina.ca Abstract.

More information

Machine Learning

Machine Learning Machine Learning 10-701 Tom M. Mitchell Machine Learning Department Carnegie Mellon University January 13, 2011 Today: The Big Picture Overfitting Review: probability Readings: Decision trees, overfiting

More information

Foundations of Artificial Intelligence

Foundations of Artificial Intelligence Foundations of Artificial Intelligence 31. Propositional Logic: DPLL Algorithm Malte Helmert and Gabriele Röger University of Basel April 24, 2017 Propositional Logic: Overview Chapter overview: propositional

More information

Machine Learning for Structured Prediction

Machine Learning for Structured Prediction Machine Learning for Structured Prediction Grzegorz Chrupa la National Centre for Language Technology School of Computing Dublin City University NCLT Seminar Grzegorz Chrupa la (DCU) Machine Learning for

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

A Probabilistic Relational Model for Characterizing Situations in Dynamic Multi-Agent Systems

A Probabilistic Relational Model for Characterizing Situations in Dynamic Multi-Agent Systems A Probabilistic Relational Model for Characterizing Situations in Dynamic Multi-Agent Systems Daniel Meyer-Delius 1, Christian Plagemann 1, Georg von Wichert 2, Wendelin Feiten 2, Gisbert Lawitzky 2, and

More information

Uncertainty. Logic and Uncertainty. Russell & Norvig. Readings: Chapter 13. One problem with logical-agent approaches: C:145 Artificial

Uncertainty. Logic and Uncertainty. Russell & Norvig. Readings: Chapter 13. One problem with logical-agent approaches: C:145 Artificial C:145 Artificial Intelligence@ Uncertainty Readings: Chapter 13 Russell & Norvig. Artificial Intelligence p.1/43 Logic and Uncertainty One problem with logical-agent approaches: Agents almost never have

More information

Fast Adaptive Algorithm for Robust Evaluation of Quality of Experience

Fast Adaptive Algorithm for Robust Evaluation of Quality of Experience Fast Adaptive Algorithm for Robust Evaluation of Quality of Experience Qianqian Xu, Ming Yan, Yuan Yao October 2014 1 Motivation Mean Opinion Score vs. Paired Comparisons Crowdsourcing Ranking on Internet

More information

Estimating query rewriting quality over LOD

Estimating query rewriting quality over LOD Semantic Web 1 (2016) 1 5 1 IOS Press Estimating query rewriting quality over LOD Editor(s): Name Surname, University, Country Solicited review(s): Name Surname, University, Country Open review(s): Name

More information

Outline. Rules of Inferences Discrete Mathematics I MATH/COSC 1056E. Example: Existence of Superman. Outline

Outline. Rules of Inferences Discrete Mathematics I MATH/COSC 1056E. Example: Existence of Superman. Outline Outline s Discrete Mathematics I MATH/COSC 1056E Julien Dompierre Department of Mathematics and Computer Science Laurentian University Sudbury, August 6, 2008 Using to Build Arguments and Quantifiers Outline

More information

Introduction to Estimation. Martina Litschmannová K210

Introduction to Estimation. Martina Litschmannová K210 Introduction to Estimation Martina Litschmannová martina.litschmannova@vsb.cz K210 Populations vs. Sample A population includes each element from the set of observations that can be made. A sample consists

More information

Mining coreference relations between formulas and text using Wikipedia

Mining coreference relations between formulas and text using Wikipedia Mining coreference relations between formulas and text using Wikipedia Minh Nghiem Quoc 1, Keisuke Yokoi 2, Yuichiroh Matsubayashi 3 Akiko Aizawa 1 2 3 1 Department of Informatics, The Graduate University

More information

DESCRIPTION LOGICS. Paula Severi. October 12, University of Leicester

DESCRIPTION LOGICS. Paula Severi. October 12, University of Leicester DESCRIPTION LOGICS Paula Severi University of Leicester October 12, 2009 Description Logics Outline Introduction: main principle, why the name description logic, application to semantic web. Syntax and

More information

Markov Networks. l Like Bayes Nets. l Graphical model that describes joint probability distribution using tables (AKA potentials)

Markov Networks. l Like Bayes Nets. l Graphical model that describes joint probability distribution using tables (AKA potentials) Markov Networks l Like Bayes Nets l Graphical model that describes joint probability distribution using tables (AKA potentials) l Nodes are random variables l Labels are outcomes over the variables Markov

More information

COMS 4771 Probabilistic Reasoning via Graphical Models. Nakul Verma

COMS 4771 Probabilistic Reasoning via Graphical Models. Nakul Verma COMS 4771 Probabilistic Reasoning via Graphical Models Nakul Verma Last time Dimensionality Reduction Linear vs non-linear Dimensionality Reduction Principal Component Analysis (PCA) Non-linear methods

More information

arxiv: v1 [cs.db] 14 May 2017

arxiv: v1 [cs.db] 14 May 2017 Discovering Multiple Truths with a Model Furong Li Xin Luna Dong Anno Langen Yang Li National University of Singapore Google Inc., Mountain View, CA, USA furongli@comp.nus.edu.sg {lunadong, arl, ngli}@google.com

More information

Lecture Notes in Machine Learning Chapter 4: Version space learning

Lecture Notes in Machine Learning Chapter 4: Version space learning Lecture Notes in Machine Learning Chapter 4: Version space learning Zdravko Markov February 17, 2004 Let us consider an example. We shall use an attribute-value language for both the examples and the hypotheses

More information

Hike: A Hybrid Human-Machine Method for Entity Alignment in Large-Scale Knowledge Bases

Hike: A Hybrid Human-Machine Method for Entity Alignment in Large-Scale Knowledge Bases Hike: A Hybrid Human-Machine Method for Entity Alignment in Large-Scale Knowledge Bases Yan Zhuang, Guoliang Li, Zhuojian Zhong, Yudian Zheng #, Jianhua Feng Tsinghua University, UC Irvine, # The University

More information

A Probabilistic Model for Canonicalizing Named Entity Mentions. Dani Yogatama Yanchuan Sim Noah A. Smith

A Probabilistic Model for Canonicalizing Named Entity Mentions. Dani Yogatama Yanchuan Sim Noah A. Smith A Probabilistic Model for Canonicalizing Named Entity Mentions Dani Yogatama Yanchuan Sim Noah A. Smith Introduction Model Experiments Conclusions Outline Introduction Model Experiments Conclusions Outline

More information

Lecture 12: Link Analysis for Web Retrieval

Lecture 12: Link Analysis for Web Retrieval Lecture 12: Link Analysis for Web Retrieval Trevor Cohn COMP90042, 2015, Semester 1 What we ll learn in this lecture The web as a graph Page-rank method for deriving the importance of pages Hubs and authorities

More information

Product rule. Chain rule

Product rule. Chain rule Probability Recap CS 188: Artificial Intelligence ayes Nets: Independence Conditional probability Product rule Chain rule, independent if and only if: and are conditionally independent given if and only

More information

Machine Learning (CS 567) Lecture 2

Machine Learning (CS 567) Lecture 2 Machine Learning (CS 567) Lecture 2 Time: T-Th 5:00pm - 6:20pm Location: GFS118 Instructor: Sofus A. Macskassy (macskass@usc.edu) Office: SAL 216 Office hours: by appointment Teaching assistant: Cheol

More information

Guest Speaker. CS 416 Artificial Intelligence. First-order logic. Diagnostic Rules. Causal Rules. Causal Rules. Page 1

Guest Speaker. CS 416 Artificial Intelligence. First-order logic. Diagnostic Rules. Causal Rules. Causal Rules. Page 1 Page 1 Guest Speaker CS 416 Artificial Intelligence Lecture 13 First-Order Logic Chapter 8 Topics in Optimal Control, Minimax Control, and Game Theory March 28 th, 2 p.m. OLS 005 Onesimo Hernandez-Lerma

More information

Overview. Machine Learning, Chapter 2: Concept Learning

Overview. Machine Learning, Chapter 2: Concept Learning Overview Concept Learning Representation Inductive Learning Hypothesis Concept Learning as Search The Structure of the Hypothesis Space Find-S Algorithm Version Space List-Eliminate Algorithm Candidate-Elimination

More information

Bayesian Networks BY: MOHAMAD ALSABBAGH

Bayesian Networks BY: MOHAMAD ALSABBAGH Bayesian Networks BY: MOHAMAD ALSABBAGH Outlines Introduction Bayes Rule Bayesian Networks (BN) Representation Size of a Bayesian Network Inference via BN BN Learning Dynamic BN Introduction Conditional

More information

Fusing Data with Correlations

Fusing Data with Correlations Fusing Data with Correlations Ravali Pochampally University of Massachusetts ravali@cs.umass.edu Alexandra Meliou University of Massachusetts ameli@cs.umass.edu Anish Das Sarma Troo.Ly Inc. anish.dassarma@gmail.com

More information