SSSW, profile picture
Uploaded bySSSW
1,572 views

Navigli sssw

Roberto Navigli gave a presentation on BabelNet, Babelfy, and related projects. BabelNet is a multilingual semantic network created by merging various knowledge resources, including WordNet, Wikipedia, Wiktionary, and other wordnets. It contains over 14 million entries in 271 languages and integrates both encyclopedic and lexicographic knowledge. BabelNet aims to provide a unified semantic representation of concepts across languages to support multilingual natural language processing applications.

Education◦
Related topics:
Semantic Web•

Embed presentation

Downloaded 19 times
Roberto Navigli Multilinguality at Your Fingertips: BabelNet, Babelfy and Beyond! Bertinoro, 8th July 2015 http://lcl.uniroma1.it ERC Starting Grant n. 259234 LIDER CSA n. 610782
10/07/2015 Simone Ponzetto Tiziano Flati Andrea Moro Daniele Vannella Taher Pilehvar Francesco Cecconi 10/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli 2 Federico Scozzafava
10/07/2015Multilingual Web Access – WWW 2015 Roberto Navigli 3
BabelNet, Babelfy, Video games with a purpose & the Wikipedia Bitaxonomy Roberto Navigli 4 It’s all about knowledge! • But can we expect computers to know? • Can’t computers just use, e.g., statistical techniques?
State-of-the-art Machine Translation • EN: These are movies in which the music genre, e.g. rock, is an important element but not necessarily central to the plot. Examples are Easy Rider (1969), The Graduate (1969), and Saturday Night Fever (1978). BabelNet, Babelfy and Beyond! Roberto Navigli
State-of-the-art Machine Translation • EN: These are movies in which the music genre, e.g. rock, is an important element but not necessarily central to the plot. Examples are Easy Rider (1969), The Graduate (1969), and Saturday Night Fever (1978). • FR: Ce sont des films dans lesquels le genre de musique, par exemple, rock, est un élément important, mais pas nécessairement au centre de l'intrigue. Les exemples sont Easy Rider (1969), The Graduate (1969), et Saturday Night Fever (1978). BabelNet, Babelfy and Beyond! Roberto Navigli
State-of-the-art Machine Translation • EN: These are movies in which the music genre, e.g. rock, is an important element but not necessarily central to the plot. Examples are Easy Rider (1969), The Graduate (1969), and Saturday Night Fever (1978). • ES: Estas son las películas en las que el género de la música, por ejemplo, roca, es un elemento importante, pero no necesariamente el centro de la trama. [...] BabelNet, Babelfy and Beyond! Roberto Navigli
State-of-the-art Machine Translation • EN: We can look at how this vast slug of molten underground rock was injected. Danger here! BabelNet, Babelfy and Beyond! Roberto Navigli
State-of-the-art Machine Translation • EN: We can look at how this vast slug of molten underground rock was injected. • FR: Nous pouvons voir comment ce vaste bouchon de rock underground fondu a été injecté. • IT: Possiamo guardare a come è stato iniettato questo vasto slug del rock underground fusa. BabelNet, Babelfy and Beyond! Roberto Navigli
10/07/2015 11 What are we talking about? A 5-year ERC Starting Grant (2011-2016) on Multilingual Word Sense Disambiguation BabelNet, Babelfy and Beyond! Roberto Navigli
INTEGRATING KNOWLEDGE [Navigli & Ponzetto, ACL 2010; Pilehvar & Navigli, ACL 2014] 10/07/2015 16BabelNet, Babelfy and Beyond! Roberto Navigli
10/07/2015 14 Key Objective 1: create knowledge for all languages Multilingual Joint Word Sense Disambiguation (MultiJEDI) WordNet MultiWordNet WOLF MCR GermaNet BalkaNet BabelNet, Babelfy and Beyond! Roberto Navigli
It all started with merging WordNet and Wikipedia [Navigli and Ponzetto, ACL 2010; AIJ 2012] • A wide-coverage multilingual semantic network including both encyclopedic (from Wikipedia) and lexicographic (from WordNet) entries Concepts from WordNetNamed Entities and specialized concepts from Wikipedia Concepts integrated from both resources 10/07/2015 15BabelNet, Babelfy and Beyond! Roberto Navigli
Creating a Multilingual Semantic Network • Start from two large complementary resources: – WordNet: full-fledged taxonomy – Wikipedia: multilingual and continuously updated {wheeled vehicle} {self-propelled vehicle} {motor vehicle} {tractor} {car,auto, automobile, machine, motorcar} {convertible} {air bag} is-a is-a is-a is-a is-a has-part {golf cart, golfcart} is-a {wagon, waggon} is-a {accelerator, accelerator pedal, gas pedal, throttle} has-part {car window} has-part {locomotive, engine, locomotive engine, railway locomotive} is-a {brake}has-part {wheel} has-part {splasher} has-part Get the best from both worlds 16BabelNet, Babelfy and Beyond! Roberto Navigli
{wheeled vehicle} {self-propelled vehicle} {motor vehicle} {tractor} {car,auto, automobile, machine, motorcar} {convertible} {air bag} is-a is-a is-a is-a is-a has-part{golf cart, golfcart} is-a {wagon, waggon} is-a {accelerator, accelerator pedal, gas pedal, throttle} has-part {car window} has-part {locomotive, engine, locomotive engine, railway locomotive} is-a {brake}has-part {wheel} has-part {splasher} has-part concepts semantic relation WordNet [Miller et al., 1990; Fellbaum, 1998] 17BabelNet, Babelfy and Beyond! Roberto Navigli
BabelNet, Babelfy, Video games with a purpose & the Wikipedia Bitaxonomy Roberto Navigli 18 • Playing with senses • Bla bla bla bla bla bla bla • Bla bla bla bla bla bla bla • Bla bla bla bla bla bla bla • Bla bla bla bla bla bla bla • Bla bla bla bla bla bla bla concepts (unspecified) semantic relation Wikipedia [The Web Community, 2001-today]
Structural Similarity with Personalized PageRank [Pilehvar and Navigli, ACL 2014] some
Structural Similarity with Personalized PageRank [Pilehvar and Navigli, ACL 2014]
To merge or not to merge? [Pilehvar and Navigli, ACL 2014] • Measure the similarity of senses of the same word (but from different resources) • If they are similar enough, merge the corresponding two concepts WordNetplant#n#1plant#n#1 21BabelNet, Babelfy and Beyond! Roberto Navigli
• We collect lexicalizations, definitions, translations, images, etc. from each of the merged resources Merging entries from different resources into BabelNet BabelNet, Babelfy and Beyond! Roberto Navigli 22 WordNet
BabelNet: concepts and semantic relations (2) • We encode knowledge as a labeled directed graph: – Each vertex is a Babel synset – Each edge is a semantic relation between synsets: • is-a (balloon is-a aircraft) • part-of (gasbag part-of balloon) • instance-of (Einstein instance-of physicist) • … • unspecified/relatedness (balloon related-to flight) balloonEN, BallonDE, aerostatoES, aerostatoIT, pallone aerostaticoIT, mongolfièreFR 3210/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
Building BabelNet: Translating Babel synsets 1. Exploiting Wikipedia interlanguage links pallone aerostatico globo aerostàtico Ballon 3310/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
Building BabelNet: Translating Babel synsets 2. Filling the lexical translation gaps using a Machine Translation system to translate the English lexicalizations of a concept • On August 27, 1783 in Paris, Franklin witnessed the world's first hydrogen [[Balloon (aircraft)|balloon]] flight. • Le 27 Août, 1783 à Paris, Franklin vu le premier vol en ballon d'hydrogène. Statistical Machine Translation 3410/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
The most frequent translation of a word in a given meaning left context term right context wikification may refer to: the… geoinformatics services' and ' wikification of GIS by the masses' the process may be called wikification (as in ... which is then called " wikification and to the related problem reason needs copyediting, wikification , reduction of POV, work on references huge amount of cleanup, wikification , etc. Version of 12 Nov 3610/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
left context term right context wikificazione potrebbe riferirsi a: il… servizi geoinformatici' e ' wikification di GIS dalle masse' il processo chiamato wikificazione (come in ... che è quindi chiamato wikificazione e al problema correlato… ragione richiede copyediting, wikification , riduzione di POV, lavoro su reference grandi quantità di pulizia, wikificazione , ecc. Versione del 12 Novembre The most frequent translation of a word in a given meaning 3710/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
left context term right context wikificazione potrebbe riferirsi a: il… servizi geoinformatici' e ' wikification di GIS dalle masse' il processo chiamato wikificazione (come in ... che è quindi chiamato wikificazione e al problema correlato… ragione richiede copyediting, wikification , riduzione di POV, lavoro su reference grandi quantità di pulizia, wikificazione , ecc. Versione del 12 Novembre The most frequent translation of a word in a given meaning 3810/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
What is BabelNet? • A merger of resources of different kinds: 10/07/2015META Prize 2015: BabelNet Roberto Navigli 39
10/07/2015 40 What is BabelNet? • A merger of resources of different kinds: – WordNet: the most popular computational lexicon of English – Open Multilingual WordNet: a collection of open wordnets – Wikipedia: the largest collaborative encyclopedia – Wikidata: the largest collaborative knowledge base – Wiktionary: the largest collaborative dictionary – OmegaWiki: a medium-size collaborative multilingual dictionary – High-quality automatic sense-based translations BabelNet, Babelfy and Beyond! Roberto Navigli
10/07/2015 41 What is BabelNet? • A merger of resources of different kinds: BabelNet, Babelfy and Beyond! Roberto Navigli
10/07/2015 42 Why do we need BabelNet? • Multilinguality: the same concept is expressed in tens of languages BabelNet, Babelfy and Beyond! Roberto Navigli
10/07/2015 43 Why do we need BabelNet? • Multilinguality: the same concept is expressed in tens of languages BabelNet, Babelfy and Beyond! Roberto Navigli
10/07/2015Multilingual Web Access – WWW 2015 Roberto Navigli 44
10/07/2015 45 Why do we need BabelNet? • Multilinguality: the same concept is expressed in tens of languages • Coverage: 271 languages and 14 million entries! BabelNet, Babelfy and Beyond! Roberto Navigli
10/07/2015Multilingual Web Access – WWW 2015 Roberto Navigli 46 Why do we need BabelNet? • Multilinguality: the same concept is expressed in tens of languages • Coverage: 271 languages and 14 million entries! • Concepts and named entities together: dictionary and encyclopedic knowledge is semantically interconnected 10/07/2015META Prize 2015: BabelNet Roberto Navigli 46
10/07/2015Multilingual Web Access – WWW 2015 Roberto Navigli 47 Why do we need BabelNet? • Multilinguality: the same concept is expressed in tens of languages • Coverage: 271 languages and 14 million entries! • Concepts and named entities together: dictionary and encyclopedic knowledge is semantically interconnected • "Dictionary of the future": semantic network structure with labeled relations, pictures, multilingual synsets 10/07/2015META Prize 2015: BabelNet Roberto Navigli 47
10/07/2015 48 Why do we need BabelNet? • Multilinguality: the same concept is expressed in tens of languages • Coverage: 271 languages and 14 million entries! • Concepts and named entities together: dictionary and encyclopedic knowledge is semantically interconnected • "Dictionary of the future": semantic network structure with labeled relations, pictures, multilingual synsets • Full-fledged taxonomy: is-a relations are available for both concepts and named entities (Wikipedia Bitaxonomy) – Bertinoro is-a town & comune – BabelNet is-a semantic network & encyclopedic dictionary – enjoyment is-a pleasure – Couldn't find a hypernym for ontology learning – Ontology engineering is-a Winchester College Ground (!!!) BabelNet, Babelfy and Beyond! Roberto Navigli
10/07/2015Multilingual Web Access – WWW 2015 Roberto Navigli 49 DON’T PANIC: this time has yet to come!!! But you are working on it, aren't you?
10/07/2015 50 Why do we need BabelNet? • Easy access: Java and HTTP RESTful APIs; SPARQL endpoint (2 billion triples) SPARQL endpoint: http://babelnet.org/sparql/ BabelNet, Babelfy and Beyond! Roberto Navigli
The core of the Linguistic Linked Open Data cloud!
10/07/2015 52 What can we do with BabelNet? • Search and translate: BabelNet, Babelfy and Beyond! Roberto Navigli
10/07/2015META Prize 2015: BabelNet Roberto Navigli 53 What can we do with BabelNet?
What can we do with BabelNet? • Explore the network: 10/07/2015META Prize 2015: BabelNet Roberto Navigli 54
We are not alone in the (resource) universe! 10/07/2015BabelNet: a Very Large Multilingual Ontology Roberto Navigli 71
We are not alone in the (resource) universe! • DBpedia [Bizer et al. 2009] - a resource obtained from structured information in Wikipedia – «Describes 3.77M things» – No dictionary side • YAGO [Suchanek et al. 2007] – «Contains 10M entities and 120M facts about these entities» – Links Wikipedia categories to WordNet synsets • MENTA [de Melo and Weikum, 2010] – A «multilingual taxonomy with 5.4M entities» • WikiNet [Nastase and Strube, 2013] – Semantic network connecting Wikipedia entities – «3M concepts and 38+M relations» • Freebase (http://freebase.com): collaborative effort – Started from Wikipedia, MusicBrainz, ChefMoz, etc. Shut down! 72BabelNet, Babelfy and Beyond! Roberto Navigli
10/07/2015BabelNet & friends Roberto Navigli 75
ADDRESSING AMBIGUITY [Moro, Raganato & Navigli, TACL 2014] BabelNet, Babelfy and Beyond! Roberto Navigli 97
Context matters!!! 10/07/2015 98BabelNet, Babelfy and Beyond! Roberto Navigli
Back to our issue: lexical ambiguity! • Thomas and Mario played as strikers in Munich. 9910/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
Word Sense Disambiguation and Entity Linking • Thomas and Mario are strikers playing in Munich Entity Linking: The task of discovering mentions of entities within a text and linking them in a knowledge base. WSD: The task aimed at assigning meanings to word occurrences within text. 10010/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
Word Sense Disambiguation in a Nutshell strikers (target word) “Thomas and Mario are strikers playing in Munich” (context) WSD system knowledge sense of target word 10110/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
Entity Linking in a Nutshell Thomas (target mention) “Thomas and Mario are strikers playing in Munich” (context) EL system Named Entity 102 knowledge 10/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
Disambiguation and Entity Linking together! BabelNet is a huge multilingual inventory for both word senses and named entities! 10310/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
Multilingual Joint Word Sense Disambiguation (MultiJEDI) Key Objective 2: use all languages to disambiguate one 10410/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
So what? 10510/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
Step 1: Find all possible meanings of words 1. Exact Matching (good for WSD, bad for EL) Thomas and Mario are strikers playing in Munich Thomas, Norman Thomas, Seth They both have Thomas as one of their lexicalizations 10810/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
Step 1: Find all possible meanings of words 2. Partial Matching (good for EL) Thomas and Mario are strikers playing in Munich Thomas, Norman Thomas, Seth Thomas Müller It has Thomas as a substring of one of its lexicalizations 10910/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
Step 1: Find all possible meanings of words “Thomas and Mario are strikers playing in Munich” Thomas (novel) Seth Thomas Thomas Müller Mario Gómez Mario (Album) Mario (Character) Striker (Movie) Striker (Video Game) striker (Sport) Munich (City) FC Bayern Munich Munich (Song) 11010/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
Step 1: Find all possible meanings of words “Thomas and Mario are strikers playing in Munich” Thomas (novel) Seth Thomas Thomas Müller Mario Gómez Mario (Album) Mario (Character) Striker (Movie) Striker (Video Game) striker (Sport) Munich (City) FC Bayern Munich Munich (Song) Ambiguity! 11110/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
Step 2: Connect all the candidate meanings Thomas and Mario are strikers playing in Munich 11310/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
Step 3: Extract a dense subgraph Thomas and Mario are strikers playing in Munich 11510/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
Step 3: Extract a dense subgraph Thomas and Mario are strikers playing in Munich 11610/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
Step 4: Select the most reliable meanings Thomas and Mario are strikers playing in Munich 11910/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
Step 4: Select the most reliable meanings “Thomas and Mario are strikers playing in Munich” Thomas (novel) Seth Thomas Thomas Müller Mario Gómez Mario (Album) Mario (Character) Striker (Movie) Striker (Video Game) striker (Sport) Munich (City) FC Bayern Munich Munich (Song) 12010/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
Experimental Results: Fine-grained (Multilingual) Disambiguation Senseval-3 SemEval-2007 task 17 SemEval-2013 task 12 12210/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
Experimental Results: KORE50, AIDA-CoNLL • Two gold-standard Entity Linking datasets: 12410/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
10/07/2015 125 What can we do with Babelfy? • Disambiguate text written in any language! BabelNet, Babelfy and Beyond! Roberto Navigli
10/07/2015 126 What can we do with Babelfy? • Disambiguate text written in any language! • Disambiguate in a language-agnostic setting! BabelNet, Babelfy and Beyond! Roberto Navigli
10/07/2015BabelNet & friends Roberto Navigli 127
Now in the Linked Data cloud… 10/07/2015 128BabelNet goes to the (Multilingual) Semantic Web Roberto Navigli
The Linguistic Linked Open Data Cloud
BabelNet in Lemon/RDF 10/07/2015Multilingual Web Access – WWW 2015 Roberto Navigli 130 Babel words lemmas and inflected words Babel senses Babel synsets • BabelNet is a proof of concept: a multilingual (271 languages) encyclopedic dictionary and a semantic network in the LLOD cloud. • Choice of model: lemon, mainly, but also SKOS (for concepts and narrower/broader relations), OWL and LexInfo (meronym/holonym relations, POS, translations) Babel gloss
Now: what can I do with my (linguistic? domain-oriented?) resource? Your (linguistic) resource 10/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
10/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli Now: what can I do with my (linguistic? domain-oriented?) resource?
10/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli Now: what can I do with my (linguistic? domain-oriented?) resource?
10/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli Now: what can I do with my (linguistic? domain-oriented?) resource?
A case study: the IATE term base • EU's inter-institutional terminology database • 1.4 million multilingual entries 13510/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
A case study: the IATE term base 13610/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
Early achievement: linking IATE to BabelNet • Goal: To automatically (and semantically) link IATE to BabelNet using a language- and resource-agnostic approach 10/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
IATE-258730 bn:00491522nCorylus maxima Early achievement: linking IATE to BabelNet 10/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
Linking "pomodoro di serra" to BabelNet • Babelfy features language-agnostic disambiguation!
Linking "pomodoro di serra" to BabelNet 10/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
Summarizing 10/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli 203
Actually there’s much much much more! But no time NOW! 20410/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli (but I'll be here till Saturday!!!)
10/07/2015BabelNet & friends Roberto Navigli 205
Thanks or… m i(grazie) 20610/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
Roberto Navigli Linguistic Computing Laboratory http://lcl.uniroma1.it

More Related Content

PDF
An introduction to Linked (Open) Data
byAli Khalili
 
PDF
What is pattern recognition (lecture 4 of 6)
byRanda Elanwar
 
PPTX
Studying Migrations Routes: New data and Tools
byIstituto nazionale di statistica
 
PPTX
Linked Open Data
byLars Marius Garshol
 
PPTX
Tutorial Cognition - Irene
bySSSW
 
PPTX
Entity linking meets Word Sense Disambiguation: a unified approach(TACL 2014)の紹介
byKoji Matsuda
 
PDF
Introduction to open data quality et
byOpen Data Support
 
PPTX
Pattern recognition
byShailesh Thakur
 
An introduction to Linked (Open) Data
byAli Khalili
 
What is pattern recognition (lecture 4 of 6)
byRanda Elanwar
 
Studying Migrations Routes: New data and Tools
byIstituto nazionale di statistica
 
Linked Open Data
byLars Marius Garshol
 
Tutorial Cognition - Irene
bySSSW
 
Entity linking meets Word Sense Disambiguation: a unified approach(TACL 2014)の紹介
byKoji Matsuda
 
Introduction to open data quality et
byOpen Data Support
 
Pattern recognition
byShailesh Thakur
 

Viewers also liked

PDF
An introduction to open data
bySally Lait
 
PDF
LOD(Linked Open Data) Recommendations
byMyungjin Lee
 
PPT
Data Classification Presentation
byDerroylo
 
PDF
Introduction to linked data
byOpen Data Support
 
PPTX
Pattern recognition and Machine Learning.
byRohit Kumar
 
PPT
Data Mining: Concepts and techniques classification _chapter 9 :advanced methods
bySalah Amean
 
PDF
Linked Open Data Principles, Technologies and Examples
byOpen Data Support
 
PPTX
Statistical classification: A review on some techniques
byGiorgos Bamparopoulos
 
PDF
Seminar(Pattern Recognition)
byanurodhsinha
 
PPTX
Chapter 4 Classification
byKhalid Elshafie
 
PPT
Pattern recognition
byArmando Vieira
 
PDF
Lecture 4 Decision Trees (2): Entropy, Information Gain, Gain Ratio
byMarina Santini
 
PPTX
Artificial intelligence Pattern recognition system
byREHMAT ULLAH
 
PDF
Introduction to pattern recognition
byLuís Gustavo Martins
 
PPTX
Pattern Recognition
byMaaz Hasan
 
PPT
EIFL 2014 - Linked Open Data
byAntoine Isaac
 
PDF
Quick Linked Data Introduction
byMichael Hausenblas
 
PDF
Linked Data Tutorial
byMichael Hausenblas
 
PDF
Open Data and Linked Data
byJames G. Boram Kim
 
PPT
Linked Open Data
byDerilinx
 
An introduction to open data
bySally Lait
 
LOD(Linked Open Data) Recommendations
byMyungjin Lee
 
Data Classification Presentation
byDerroylo
 
Introduction to linked data
byOpen Data Support
 
Pattern recognition and Machine Learning.
byRohit Kumar
 
Data Mining: Concepts and techniques classification _chapter 9 :advanced methods
bySalah Amean
 
Linked Open Data Principles, Technologies and Examples
byOpen Data Support
 
Statistical classification: A review on some techniques
byGiorgos Bamparopoulos
 
Seminar(Pattern Recognition)
byanurodhsinha
 
Chapter 4 Classification
byKhalid Elshafie
 
Pattern recognition
byArmando Vieira
 
Lecture 4 Decision Trees (2): Entropy, Information Gain, Gain Ratio
byMarina Santini
 
Artificial intelligence Pattern recognition system
byREHMAT ULLAH
 
Introduction to pattern recognition
byLuís Gustavo Martins
 
Pattern Recognition
byMaaz Hasan
 
EIFL 2014 - Linked Open Data
byAntoine Isaac
 
Quick Linked Data Introduction
byMichael Hausenblas
 
Linked Data Tutorial
byMichael Hausenblas
 
Open Data and Linked Data
byJames G. Boram Kim
 
Linked Open Data
byDerilinx
 

Similar to Navigli sssw

PDF
Roberto Navigli - From Text to Concepts and Back: Going Multilingual with Bab...
byMeetupDataScienceRoma
 
PPTX
Semantic Web in the Plateau of Productivity
byIoannis Stavrakantonakis
 
PDF
Semantic Technologies for Cultural Heritage
byVladimir Alexiev, PhD, PMP
 
PPT
LIS 653 Posters
byPrattSILS
 
PDF
Semantics-aware Content-based Recommender Systems
byPasquale Lops
 
PPT
Ontologies for multimedia: the Semantic Culture Web
byGuus Schreiber
 
PPTX
BabelNet 3.0
byShrikrishna Parab
 
PPT
Ontology Poster
byPrattSILS
 
PPTX
Introduction to development of lexical databases
byMuhammad Shoaib Chaudhary
 
PDF
Books and Webs: Pulling the Down Rows
byPeter Brantley
 
PPT
Porting Library Vocabularies to the Semantic Web - IFLA 2010
byBernard Vatant
 
PDF
Semantic web and information graph
byChao-Hsuan Shen
 
PPT
Introduction
bysriniefs
 
PPT
Semantic Search overview at SSSW 2012
byPeter Mika
 
PDF
Interactive Video Search and Browsing Systems
byMedia Integration and Communication Center
 
PDF
Interactive Video Search and Browsing Systems
byAndrea Ferracani
 
PDF
The Web: history - now - future
byKnowledge Hives
 
PDF
Overview of-semantic-technologies-and-ontologies
byAndrea Westerinen
 
PDF
Fact forge aimsa2012
byMariana Damova, Ph.D
 
PDF
The promise of structured data and blocks
byOmar Reiss
 
Roberto Navigli - From Text to Concepts and Back: Going Multilingual with Bab...
byMeetupDataScienceRoma
 
Semantic Web in the Plateau of Productivity
byIoannis Stavrakantonakis
 
Semantic Technologies for Cultural Heritage
byVladimir Alexiev, PhD, PMP
 
LIS 653 Posters
byPrattSILS
 
Semantics-aware Content-based Recommender Systems
byPasquale Lops
 
Ontologies for multimedia: the Semantic Culture Web
byGuus Schreiber
 
BabelNet 3.0
byShrikrishna Parab
 
Ontology Poster
byPrattSILS
 
Introduction to development of lexical databases
byMuhammad Shoaib Chaudhary
 
Books and Webs: Pulling the Down Rows
byPeter Brantley
 
Porting Library Vocabularies to the Semantic Web - IFLA 2010
byBernard Vatant
 
Semantic web and information graph
byChao-Hsuan Shen
 
Introduction
bysriniefs
 
Semantic Search overview at SSSW 2012
byPeter Mika
 
Interactive Video Search and Browsing Systems
byMedia Integration and Communication Center
 
Interactive Video Search and Browsing Systems
byAndrea Ferracani
 
The Web: history - now - future
byKnowledge Hives
 
Overview of-semantic-technologies-and-ontologies
byAndrea Westerinen
 
Fact forge aimsa2012
byMariana Damova, Ph.D
 
The promise of structured data and blocks
byOmar Reiss
 

Recently uploaded

DOCX
Mobile applications Devlopment ReTest year 2025-2026
byDr. Mazin Mohamed alkathiri
 
PPTX
15 December 2025 Education for human flourishing Michael Stevenson .pptx
byEduSkills OECD
 
PPTX
Pain. definition, causes, factor influencing pain & pain assessment.pptx
byPompi Begum
 
PPTX
Limpitlaw "Licensing: From Mindset to Milestones"
byNational Information Standards Organization (NISO)
 
PDF
UKSG Forum 2025 - They asked for everything - The Case of the Systematic Revi...
byUKSG: connecting the knowledge community
 
PPTX
2025-2026 History in your Hands Class 4 December 2025 January 2026 .pptx
byEilsONeill
 
PDF
NAVIGATE PHARMACY CAREER OPPORTUNITIES.pdf
byMd. Zakaria Faruki
 
PDF
Blue / Green: Troop Leading Procedure (TLP) Overview.pdf
byBlue / Green Training
 
PDF
Current Electricity for first year physiotherapy
byGanesh Jadhav
 
PPTX
How to use search_read method in Odoo 18
byCeline George
 
PPTX
York "Collaboration for Research Support at U-M Library"
byNational Information Standards Organization (NISO)
 
PDF
Projecte de la porta de primer B: L'antic Egipte
byeduardrubio1
 
PDF
All Students Workshop 25 Yoga Wellness by LDMMIA
by©LDMMIA, ©Reiki Yoga
 
PPTX
The Cell & Cell Cycle-detailed structure and function of organelles.pptx
byAshish Umale
 
PPTX
Searching in PubMed andCochrane_Practical Presentation.pptx
bySystematic Reviews Network (SRN)
 
PDF
Projecte de la porta de la classe de primer A: Mar i cel.
byeduardrubio1
 
PDF
DHA/HAAD/MOH/DOH OPTOMETRY MCQ PYQ. .pdf
byAnmol Singh
 
PDF
Unit-III pdf (Basic listening Skill, Effective Writing Communication & Writin...
bySayyadTarannum
 
PPTX
ICH Harmonization A Global Pathway to Unified Drug Regulation.pptx
byDr. Smita Kumbhar
 
PPTX
Semester 6 UNIT 2 Dislocation of hip.pptx
bysachin7989
 
Mobile applications Devlopment ReTest year 2025-2026
byDr. Mazin Mohamed alkathiri
 
15 December 2025 Education for human flourishing Michael Stevenson .pptx
byEduSkills OECD
 
Pain. definition, causes, factor influencing pain & pain assessment.pptx
byPompi Begum
 
Limpitlaw "Licensing: From Mindset to Milestones"
byNational Information Standards Organization (NISO)
 
UKSG Forum 2025 - They asked for everything - The Case of the Systematic Revi...
byUKSG: connecting the knowledge community
 
2025-2026 History in your Hands Class 4 December 2025 January 2026 .pptx
byEilsONeill
 
NAVIGATE PHARMACY CAREER OPPORTUNITIES.pdf
byMd. Zakaria Faruki
 
Blue / Green: Troop Leading Procedure (TLP) Overview.pdf
byBlue / Green Training
 
Current Electricity for first year physiotherapy
byGanesh Jadhav
 
How to use search_read method in Odoo 18
byCeline George
 
York "Collaboration for Research Support at U-M Library"
byNational Information Standards Organization (NISO)
 
Projecte de la porta de primer B: L'antic Egipte
byeduardrubio1
 
All Students Workshop 25 Yoga Wellness by LDMMIA
by©LDMMIA, ©Reiki Yoga
 
The Cell & Cell Cycle-detailed structure and function of organelles.pptx
byAshish Umale
 
Searching in PubMed andCochrane_Practical Presentation.pptx
bySystematic Reviews Network (SRN)
 
Projecte de la porta de la classe de primer A: Mar i cel.
byeduardrubio1
 
DHA/HAAD/MOH/DOH OPTOMETRY MCQ PYQ. .pdf
byAnmol Singh
 
Unit-III pdf (Basic listening Skill, Effective Writing Communication & Writin...
bySayyadTarannum
 
ICH Harmonization A Global Pathway to Unified Drug Regulation.pptx
byDr. Smita Kumbhar
 
Semester 6 UNIT 2 Dislocation of hip.pptx
bysachin7989
 

Navigli sssw

  • 1.
    Roberto Navigli Multilinguality atYour Fingertips: BabelNet, Babelfy and Beyond! Bertinoro, 8th July 2015 http://lcl.uniroma1.it ERC Starting Grant n. 259234 LIDER CSA n. 610782
  • 2.
  • 3.
    10/07/2015Multilingual Web Access– WWW 2015 Roberto Navigli 3
  • 4.
    BabelNet, Babelfy, Videogames with a purpose & the Wikipedia Bitaxonomy Roberto Navigli 4 It’s all about knowledge! • But can we expect computers to know? • Can’t computers just use, e.g., statistical techniques?
  • 5.
    State-of-the-art Machine Translation •EN: These are movies in which the music genre, e.g. rock, is an important element but not necessarily central to the plot. Examples are Easy Rider (1969), The Graduate (1969), and Saturday Night Fever (1978). BabelNet, Babelfy and Beyond! Roberto Navigli
  • 6.
    State-of-the-art Machine Translation •EN: These are movies in which the music genre, e.g. rock, is an important element but not necessarily central to the plot. Examples are Easy Rider (1969), The Graduate (1969), and Saturday Night Fever (1978). • FR: Ce sont des films dans lesquels le genre de musique, par exemple, rock, est un élément important, mais pas nécessairement au centre de l'intrigue. Les exemples sont Easy Rider (1969), The Graduate (1969), et Saturday Night Fever (1978). BabelNet, Babelfy and Beyond! Roberto Navigli
  • 7.
    State-of-the-art Machine Translation •EN: These are movies in which the music genre, e.g. rock, is an important element but not necessarily central to the plot. Examples are Easy Rider (1969), The Graduate (1969), and Saturday Night Fever (1978). • ES: Estas son las películas en las que el género de la música, por ejemplo, roca, es un elemento importante, pero no necesariamente el centro de la trama. [...] BabelNet, Babelfy and Beyond! Roberto Navigli
  • 8.
    State-of-the-art Machine Translation •EN: We can look at how this vast slug of molten underground rock was injected. Danger here! BabelNet, Babelfy and Beyond! Roberto Navigli
  • 9.
    State-of-the-art Machine Translation •EN: We can look at how this vast slug of molten underground rock was injected. • FR: Nous pouvons voir comment ce vaste bouchon de rock underground fondu a été injecté. • IT: Possiamo guardare a come è stato iniettato questo vasto slug del rock underground fusa. BabelNet, Babelfy and Beyond! Roberto Navigli
  • 10.
    10/07/2015 11 What arewe talking about? A 5-year ERC Starting Grant (2011-2016) on Multilingual Word Sense Disambiguation BabelNet, Babelfy and Beyond! Roberto Navigli
  • 11.
    INTEGRATING KNOWLEDGE [Navigli & Ponzetto,ACL 2010; Pilehvar & Navigli, ACL 2014] 10/07/2015 16BabelNet, Babelfy and Beyond! Roberto Navigli
  • 12.
    10/07/2015 14 Key Objective1: create knowledge for all languages Multilingual Joint Word Sense Disambiguation (MultiJEDI) WordNet MultiWordNet WOLF MCR GermaNet BalkaNet BabelNet, Babelfy and Beyond! Roberto Navigli
  • 13.
    It all startedwith merging WordNet and Wikipedia [Navigli and Ponzetto, ACL 2010; AIJ 2012] • A wide-coverage multilingual semantic network including both encyclopedic (from Wikipedia) and lexicographic (from WordNet) entries Concepts from WordNetNamed Entities and specialized concepts from Wikipedia Concepts integrated from both resources 10/07/2015 15BabelNet, Babelfy and Beyond! Roberto Navigli
  • 14.
    Creating a MultilingualSemantic Network • Start from two large complementary resources: – WordNet: full-fledged taxonomy – Wikipedia: multilingual and continuously updated {wheeled vehicle} {self-propelled vehicle} {motor vehicle} {tractor} {car,auto, automobile, machine, motorcar} {convertible} {air bag} is-a is-a is-a is-a is-a has-part {golf cart, golfcart} is-a {wagon, waggon} is-a {accelerator, accelerator pedal, gas pedal, throttle} has-part {car window} has-part {locomotive, engine, locomotive engine, railway locomotive} is-a {brake}has-part {wheel} has-part {splasher} has-part Get the best from both worlds 16BabelNet, Babelfy and Beyond! Roberto Navigli
  • 15.
    {wheeled vehicle} {self-propelled vehicle} {motorvehicle} {tractor} {car,auto, automobile, machine, motorcar} {convertible} {air bag} is-a is-a is-a is-a is-a has-part{golf cart, golfcart} is-a {wagon, waggon} is-a {accelerator, accelerator pedal, gas pedal, throttle} has-part {car window} has-part {locomotive, engine, locomotive engine, railway locomotive} is-a {brake}has-part {wheel} has-part {splasher} has-part concepts semantic relation WordNet [Miller et al., 1990; Fellbaum, 1998] 17BabelNet, Babelfy and Beyond! Roberto Navigli
  • 16.
    BabelNet, Babelfy, Videogames with a purpose & the Wikipedia Bitaxonomy Roberto Navigli 18 • Playing with senses • Bla bla bla bla bla bla bla • Bla bla bla bla bla bla bla • Bla bla bla bla bla bla bla • Bla bla bla bla bla bla bla • Bla bla bla bla bla bla bla concepts (unspecified) semantic relation Wikipedia [The Web Community, 2001-today]
  • 17.
    Structural Similarity withPersonalized PageRank [Pilehvar and Navigli, ACL 2014] some
  • 18.
    Structural Similarity withPersonalized PageRank [Pilehvar and Navigli, ACL 2014]
  • 19.
    To merge ornot to merge? [Pilehvar and Navigli, ACL 2014] • Measure the similarity of senses of the same word (but from different resources) • If they are similar enough, merge the corresponding two concepts WordNetplant#n#1plant#n#1 21BabelNet, Babelfy and Beyond! Roberto Navigli
  • 20.
    • We collectlexicalizations, definitions, translations, images, etc. from each of the merged resources Merging entries from different resources into BabelNet BabelNet, Babelfy and Beyond! Roberto Navigli 22 WordNet
  • 21.
    BabelNet: concepts andsemantic relations (2) • We encode knowledge as a labeled directed graph: – Each vertex is a Babel synset – Each edge is a semantic relation between synsets: • is-a (balloon is-a aircraft) • part-of (gasbag part-of balloon) • instance-of (Einstein instance-of physicist) • … • unspecified/relatedness (balloon related-to flight) balloonEN, BallonDE, aerostatoES, aerostatoIT, pallone aerostaticoIT, mongolfièreFR 3210/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 22.
    Building BabelNet: TranslatingBabel synsets 1. Exploiting Wikipedia interlanguage links pallone aerostatico globo aerostàtico Ballon 3310/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 23.
    Building BabelNet: TranslatingBabel synsets 2. Filling the lexical translation gaps using a Machine Translation system to translate the English lexicalizations of a concept • On August 27, 1783 in Paris, Franklin witnessed the world's first hydrogen [[Balloon (aircraft)|balloon]] flight. • Le 27 Août, 1783 à Paris, Franklin vu le premier vol en ballon d'hydrogène. Statistical Machine Translation 3410/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 24.
    The most frequenttranslation of a word in a given meaning left context term right context wikification may refer to: the… geoinformatics services' and ' wikification of GIS by the masses' the process may be called wikification (as in ... which is then called " wikification and to the related problem reason needs copyediting, wikification , reduction of POV, work on references huge amount of cleanup, wikification , etc. Version of 12 Nov 3610/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 25.
    left context termright context wikificazione potrebbe riferirsi a: il… servizi geoinformatici' e ' wikification di GIS dalle masse' il processo chiamato wikificazione (come in ... che è quindi chiamato wikificazione e al problema correlato… ragione richiede copyediting, wikification , riduzione di POV, lavoro su reference grandi quantità di pulizia, wikificazione , ecc. Versione del 12 Novembre The most frequent translation of a word in a given meaning 3710/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 26.
    left context termright context wikificazione potrebbe riferirsi a: il… servizi geoinformatici' e ' wikification di GIS dalle masse' il processo chiamato wikificazione (come in ... che è quindi chiamato wikificazione e al problema correlato… ragione richiede copyediting, wikification , riduzione di POV, lavoro su reference grandi quantità di pulizia, wikificazione , ecc. Versione del 12 Novembre The most frequent translation of a word in a given meaning 3810/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 27.
    What is BabelNet? •A merger of resources of different kinds: 10/07/2015META Prize 2015: BabelNet Roberto Navigli 39
  • 28.
    10/07/2015 40 What isBabelNet? • A merger of resources of different kinds: – WordNet: the most popular computational lexicon of English – Open Multilingual WordNet: a collection of open wordnets – Wikipedia: the largest collaborative encyclopedia – Wikidata: the largest collaborative knowledge base – Wiktionary: the largest collaborative dictionary – OmegaWiki: a medium-size collaborative multilingual dictionary – High-quality automatic sense-based translations BabelNet, Babelfy and Beyond! Roberto Navigli
  • 29.
    10/07/2015 41 What isBabelNet? • A merger of resources of different kinds: BabelNet, Babelfy and Beyond! Roberto Navigli
  • 30.
    10/07/2015 42 Why dowe need BabelNet? • Multilinguality: the same concept is expressed in tens of languages BabelNet, Babelfy and Beyond! Roberto Navigli
  • 31.
    10/07/2015 43 Why dowe need BabelNet? • Multilinguality: the same concept is expressed in tens of languages BabelNet, Babelfy and Beyond! Roberto Navigli
  • 32.
    10/07/2015Multilingual Web Access– WWW 2015 Roberto Navigli 44
  • 33.
    10/07/2015 45 Why dowe need BabelNet? • Multilinguality: the same concept is expressed in tens of languages • Coverage: 271 languages and 14 million entries! BabelNet, Babelfy and Beyond! Roberto Navigli
  • 34.
    10/07/2015Multilingual Web Access– WWW 2015 Roberto Navigli 46 Why do we need BabelNet? • Multilinguality: the same concept is expressed in tens of languages • Coverage: 271 languages and 14 million entries! • Concepts and named entities together: dictionary and encyclopedic knowledge is semantically interconnected 10/07/2015META Prize 2015: BabelNet Roberto Navigli 46
  • 35.
    10/07/2015Multilingual Web Access– WWW 2015 Roberto Navigli 47 Why do we need BabelNet? • Multilinguality: the same concept is expressed in tens of languages • Coverage: 271 languages and 14 million entries! • Concepts and named entities together: dictionary and encyclopedic knowledge is semantically interconnected • "Dictionary of the future": semantic network structure with labeled relations, pictures, multilingual synsets 10/07/2015META Prize 2015: BabelNet Roberto Navigli 47
  • 36.
    10/07/2015 48 Why dowe need BabelNet? • Multilinguality: the same concept is expressed in tens of languages • Coverage: 271 languages and 14 million entries! • Concepts and named entities together: dictionary and encyclopedic knowledge is semantically interconnected • "Dictionary of the future": semantic network structure with labeled relations, pictures, multilingual synsets • Full-fledged taxonomy: is-a relations are available for both concepts and named entities (Wikipedia Bitaxonomy) – Bertinoro is-a town & comune – BabelNet is-a semantic network & encyclopedic dictionary – enjoyment is-a pleasure – Couldn't find a hypernym for ontology learning – Ontology engineering is-a Winchester College Ground (!!!) BabelNet, Babelfy and Beyond! Roberto Navigli
  • 37.
    10/07/2015Multilingual Web Access– WWW 2015 Roberto Navigli 49 DON’T PANIC: this time has yet to come!!! But you are working on it, aren't you?
  • 38.
    10/07/2015 50 Why dowe need BabelNet? • Easy access: Java and HTTP RESTful APIs; SPARQL endpoint (2 billion triples) SPARQL endpoint: http://babelnet.org/sparql/ BabelNet, Babelfy and Beyond! Roberto Navigli
  • 39.
    The core ofthe Linguistic Linked Open Data cloud!
  • 40.
    10/07/2015 52 What canwe do with BabelNet? • Search and translate: BabelNet, Babelfy and Beyond! Roberto Navigli
  • 41.
    10/07/2015META Prize 2015:BabelNet Roberto Navigli 53 What can we do with BabelNet?
  • 42.
    What can wedo with BabelNet? • Explore the network: 10/07/2015META Prize 2015: BabelNet Roberto Navigli 54
  • 43.
    We are notalone in the (resource) universe! 10/07/2015BabelNet: a Very Large Multilingual Ontology Roberto Navigli 71
  • 44.
    We are notalone in the (resource) universe! • DBpedia [Bizer et al. 2009] - a resource obtained from structured information in Wikipedia – «Describes 3.77M things» – No dictionary side • YAGO [Suchanek et al. 2007] – «Contains 10M entities and 120M facts about these entities» – Links Wikipedia categories to WordNet synsets • MENTA [de Melo and Weikum, 2010] – A «multilingual taxonomy with 5.4M entities» • WikiNet [Nastase and Strube, 2013] – Semantic network connecting Wikipedia entities – «3M concepts and 38+M relations» • Freebase (http://freebase.com): collaborative effort – Started from Wikipedia, MusicBrainz, ChefMoz, etc. Shut down! 72BabelNet, Babelfy and Beyond! Roberto Navigli
  • 45.
  • 46.
    ADDRESSING AMBIGUITY [Moro, Raganato &Navigli, TACL 2014] BabelNet, Babelfy and Beyond! Roberto Navigli 97
  • 47.
    Context matters!!! 10/07/2015 98BabelNet,Babelfy and Beyond! Roberto Navigli
  • 48.
    Back to ourissue: lexical ambiguity! • Thomas and Mario played as strikers in Munich. 9910/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 49.
    Word Sense Disambiguationand Entity Linking • Thomas and Mario are strikers playing in Munich Entity Linking: The task of discovering mentions of entities within a text and linking them in a knowledge base. WSD: The task aimed at assigning meanings to word occurrences within text. 10010/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 50.
    Word Sense Disambiguationin a Nutshell strikers (target word) “Thomas and Mario are strikers playing in Munich” (context) WSD system knowledge sense of target word 10110/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 51.
    Entity Linking ina Nutshell Thomas (target mention) “Thomas and Mario are strikers playing in Munich” (context) EL system Named Entity 102 knowledge 10/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 52.
    Disambiguation and EntityLinking together! BabelNet is a huge multilingual inventory for both word senses and named entities! 10310/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 53.
    Multilingual Joint WordSense Disambiguation (MultiJEDI) Key Objective 2: use all languages to disambiguate one 10410/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 54.
    So what? 10510/07/2015BabelNet, Babelfyand Beyond! Roberto Navigli
  • 55.
    Step 1: Findall possible meanings of words 1. Exact Matching (good for WSD, bad for EL) Thomas and Mario are strikers playing in Munich Thomas, Norman Thomas, Seth They both have Thomas as one of their lexicalizations 10810/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 56.
    Step 1: Findall possible meanings of words 2. Partial Matching (good for EL) Thomas and Mario are strikers playing in Munich Thomas, Norman Thomas, Seth Thomas Müller It has Thomas as a substring of one of its lexicalizations 10910/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 57.
    Step 1: Findall possible meanings of words “Thomas and Mario are strikers playing in Munich” Thomas (novel) Seth Thomas Thomas Müller Mario Gómez Mario (Album) Mario (Character) Striker (Movie) Striker (Video Game) striker (Sport) Munich (City) FC Bayern Munich Munich (Song) 11010/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 58.
    Step 1: Findall possible meanings of words “Thomas and Mario are strikers playing in Munich” Thomas (novel) Seth Thomas Thomas Müller Mario Gómez Mario (Album) Mario (Character) Striker (Movie) Striker (Video Game) striker (Sport) Munich (City) FC Bayern Munich Munich (Song) Ambiguity! 11110/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 59.
    Step 2: Connectall the candidate meanings Thomas and Mario are strikers playing in Munich 11310/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 60.
    Step 3: Extracta dense subgraph Thomas and Mario are strikers playing in Munich 11510/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 61.
    Step 3: Extracta dense subgraph Thomas and Mario are strikers playing in Munich 11610/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 62.
    Step 4: Selectthe most reliable meanings Thomas and Mario are strikers playing in Munich 11910/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 63.
    Step 4: Selectthe most reliable meanings “Thomas and Mario are strikers playing in Munich” Thomas (novel) Seth Thomas Thomas Müller Mario Gómez Mario (Album) Mario (Character) Striker (Movie) Striker (Video Game) striker (Sport) Munich (City) FC Bayern Munich Munich (Song) 12010/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 65.
    Experimental Results: Fine-grained (Multilingual)Disambiguation Senseval-3 SemEval-2007 task 17 SemEval-2013 task 12 12210/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 66.
    Experimental Results: KORE50,AIDA-CoNLL • Two gold-standard Entity Linking datasets: 12410/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 67.
    10/07/2015 125 What canwe do with Babelfy? • Disambiguate text written in any language! BabelNet, Babelfy and Beyond! Roberto Navigli
  • 68.
    10/07/2015 126 What canwe do with Babelfy? • Disambiguate text written in any language! • Disambiguate in a language-agnostic setting! BabelNet, Babelfy and Beyond! Roberto Navigli
  • 69.
  • 70.
    Now in theLinked Data cloud… 10/07/2015 128BabelNet goes to the (Multilingual) Semantic Web Roberto Navigli
  • 71.
  • 72.
    BabelNet in Lemon/RDF 10/07/2015MultilingualWeb Access – WWW 2015 Roberto Navigli 130 Babel words lemmas and inflected words Babel senses Babel synsets • BabelNet is a proof of concept: a multilingual (271 languages) encyclopedic dictionary and a semantic network in the LLOD cloud. • Choice of model: lemon, mainly, but also SKOS (for concepts and narrower/broader relations), OWL and LexInfo (meronym/holonym relations, POS, translations) Babel gloss
  • 73.
    Now: what canI do with my (linguistic? domain-oriented?) resource? Your (linguistic) resource 10/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 74.
    10/07/2015BabelNet, Babelfy andBeyond! Roberto Navigli Now: what can I do with my (linguistic? domain-oriented?) resource?
  • 75.
    10/07/2015BabelNet, Babelfy andBeyond! Roberto Navigli Now: what can I do with my (linguistic? domain-oriented?) resource?
  • 76.
    10/07/2015BabelNet, Babelfy andBeyond! Roberto Navigli Now: what can I do with my (linguistic? domain-oriented?) resource?
  • 77.
    A case study:the IATE term base • EU's inter-institutional terminology database • 1.4 million multilingual entries 13510/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 78.
    A case study:the IATE term base 13610/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 79.
    Early achievement: linkingIATE to BabelNet • Goal: To automatically (and semantically) link IATE to BabelNet using a language- and resource-agnostic approach 10/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 80.
    IATE-258730 bn:00491522nCorylus maxima Earlyachievement: linking IATE to BabelNet 10/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 81.
    Linking "pomodoro diserra" to BabelNet • Babelfy features language-agnostic disambiguation!
  • 82.
    Linking "pomodoro diserra" to BabelNet 10/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli
  • 83.
  • 84.
    Actually there’s muchmuch much more! But no time NOW! 20410/07/2015BabelNet, Babelfy and Beyond! Roberto Navigli (but I'll be here till Saturday!!!)
  • 85.
  • 86.
    Thanks or… m i(grazie) 20610/07/2015BabelNet,Babelfy and Beyond! Roberto Navigli
  • 87.
    Roberto Navigli Linguistic ComputingLaboratory http://lcl.uniroma1.it