Expanding the YAGO knowledge base Rebele The YAGO knowledge base Using YAGO for the Humanities Adding Words to Expanding the YAGO knowledge base Regexes Answering Queries with Unix Shell Thomas Rebele Conclusion Télécom ParisTech 2018-07-05 1/50
Expanding the What is a knowledge base? YAGO knowledge base Rebele The YAGO knowledge base What is a knowledge base? What is YAGO? Accuracy Using YAGO for the Humanities Adding Words to Regexes married Answering Queries with Unix Albert Einstein Shell Mileva Mari´ c Conclusion 2/50
Expanding the What is a knowledge base? YAGO knowledge base Rebele The YAGO knowledge base What is a knowledge base? What is YAGO? Accuracy Using YAGO for the Humanities has advisor Adding Words to Alfred Kleiner Regexes married Answering Queries with Unix Albert Einstein Shell Mileva Mari´ c Conclusion 2/50
Expanding the What is a knowledge base? YAGO knowledge base Rebele The YAGO knowledge base What is a knowledge base? What is YAGO? Accuracy Using YAGO for the Humanities has advisor Adding Words to Alfred Kleiner Regexes married Answering won prize Queries with Unix Albert Einstein Shell Mileva Mari´ c Conclusion Nobel Prize in Physics 2/50
Expanding the What is a knowledge base? YAGO knowledge base Rebele The YAGO knowledge base What is a knowledge base? What is YAGO? Accuracy Using YAGO for the Humanities has advisor Adding Words to Alfred Kleiner Regexes married Answering won prize Queries with Unix Albert Einstein Shell Mileva Mari´ c Conclusion Nobel Prize in Physics Applications of knowledge bases ◮ question answering ◮ semantic search ◮ text analysis ◮ machine translation 2/50
Expanding the What is YAGO? YAGO knowledge base Rebele ◮ knowledge base with 10 million entities and > 210 million facts The YAGO knowledge base ◮ automatically extracted from Wikipedia, Wordnet, and Geonames What is a knowledge base? ◮ multilingual facts from 10 languages What is YAGO? ◮ focus on precision Accuracy ◮ developed by Max-Planck Institute for Informatics and Télécom Using YAGO for the Humanities ParisTech Adding Words to Regexes Answering Queries with Unix Shell Conclusion 3/50
Expanding the What is YAGO? YAGO knowledge base Rebele The YAGO knowledge base What is a knowledge base? What is YAGO? Accuracy Using YAGO for the Humanities Adding Words to Regexes ◮ I joined the project in 2015 Answering ◮ coordinated / contributed to the evaluation Queries with Unix Shell ◮ maintenance, participating in open source release Conclusion ◮ development 4/50
Expanding the What is YAGO? YAGO knowledge base Rebele The YAGO knowledge base What is a knowledge base? What is YAGO? Accuracy Using YAGO for the Humanities Adding Words to <Albert_Einstein> Regexes Answering Queries with Unix Shell married n Conclusion o w r o s i v d a Mileva Nobel Prize Alfred Kleiner 5/50
Expanding the Accuracy YAGO knowledge base Rebele The YAGO knowledge base What is a knowledge base? What is YAGO? Accuracy Using YAGO for the Humanities Adding Words to Regexes Answering Queries with Unix Shell Conclusion Figure: Screenshot of evaluation result ◮ 2 months evaluation, 15 participants ◮ evaluated 4412 facts of 76 relations (with 60m total facts) ◮ 98% facts of the sample were correct ◮ Wilson center: 95%, interval width: 4.2% 6/50 Now that we have this knowledge base, what can we do with it?
Expanding the Using YAGO for the Humanities: Related Work YAGO knowledge base Rebele The YAGO knowledge base Using YAGO for the Humanities Similar studies using Semantic Web for Digital Humanities Related Work ◮ [Schich et al., 2014]: about 150,000 people Extensions Birth and Death Dates ◮ [de la Croix et al., 2015]: about 300,000 people Place of residence Gender ◮ [Gergaud et al., 2017]: about 1,100,000 people Evaluation Life expectancy over time Births per month Relative population size Summary These studies are only about few people. Can we do better with YAGO? Adding Words to Regexes Answering Queries with Unix Shell Conclusion 7/50
Expanding the Using YAGO for the Humanities: Related Work YAGO knowledge base Rebele The YAGO knowledge base Using YAGO for the Humanities Similar studies using Semantic Web for Digital Humanities Related Work ◮ [Schich et al., 2014]: about 150,000 people Extensions Birth and Death Dates ◮ [de la Croix et al., 2015]: about 300,000 people Place of residence Gender ◮ [Gergaud et al., 2017]: about 1,100,000 people Evaluation Life expectancy over time Births per month Relative population size Summary These studies are only about few people. Can we do better with YAGO? Adding Words to Regexes Answering Queries with Unix Shell Conclusion YAGO has 2,200,000 people, but, e.g., locations only for 700,000 people How can we make YAGO more complete? 7/50
Expanding the Using YAGO for the Humanities: Birth and Death Dates YAGO knowledge base Rebele The YAGO Previous algorithm : knowledge base Using YAGO for the Humanities Plato Related Work Extensions Plato Plato was a philoso- Birth and Death Dates pher. He founded the Place of residence Academy in Athens. He Gender laid the foundation for Evaluation Languages philosophy. Life expectancy over time English Births per month Birth 428 or 427 BC German Death 348 BC Relative population size French Summary Categories: 420s BC births | 340s BC deaths | Greek philosoph | Greek Adding Words to Regexes male wrestler | Austrian writer Answering Queries with Unix Shell Extracted birth dates Conclusion 428 n O n r o b - 427 - 42# 8/50
Expanding the Using YAGO for the Humanities: Birth and Death Dates YAGO knowledge base Rebele The YAGO Previous algorithm : knowledge base Using YAGO for the Humanities Plato Related Work Extensions Plato Plato was a philoso- Birth and Death Dates pher. He founded the Place of residence Academy in Athens. He Gender laid the foundation for Evaluation Languages philosophy. Life expectancy over time English Births per month Birth 428 or 427 BC German Death 348 BC Relative population size French Summary Categories: 420s BC births | 340s BC deaths | Greek philosoph | Greek Adding Words to Regexes male wrestler | Austrian writer Answering Queries with Unix Shell Extracted birth dates Conclusion 428 n O n r o b - 427 - 42# 8/50
Expanding the Using YAGO for the Humanities: Birth and Death Dates YAGO knowledge base Rebele New algorithm: filtering with category dates The YAGO knowledge base Using YAGO for Plato the Humanities Related Work Plato Plato was a philoso- Extensions pher. He founded the Birth and Death Dates Academy in Athens. He Place of residence laid the foundation for Gender Languages philosophy. Evaluation Life expectancy over English time Birth 428 or 427 BC German Births per month Death 348 BC Relative population French size Categories: 420s BC births | 340s BC Summary deaths | Greek philosoph | Greek Adding Words to male wrestler | Austrian writer Regexes Answering Queries with Unix bornOn (infobox) Shell 428 Conclusion - 427 bornOn (category) ? 9/50 - 42#
Expanding the Using YAGO for the Humanities: Birth and Death Dates YAGO knowledge base Rebele New algorithm: filtering with category dates The YAGO knowledge base Using YAGO for Plato the Humanities Related Work Plato Plato was a philoso- Extensions pher. He founded the Birth and Death Dates Academy in Athens. He Place of residence laid the foundation for Gender Languages philosophy. Evaluation Life expectancy over English time Birth 428 or 427 BC German Births per month Death 348 BC Relative population French size Categories: 420s BC births | 340s BC Summary deaths | Greek philosoph | Greek Adding Words to male wrestler | Austrian writer Regexes Answering Queries with Unix bornOn (infobox) Shell 428 Conclusion - 427 bornOn (category) ? 9/50 - 42#
Expanding the Using YAGO for the Humanities: Birth and Death Dates YAGO knowledge base Rebele New algorithm: filtering with category dates The YAGO knowledge base Using YAGO for Plato the Humanities Related Work Plato Plato was a philoso- Extensions pher. He founded the Birth and Death Dates Academy in Athens. He Place of residence laid the foundation for Gender Languages philosophy. Evaluation Life expectancy over English time Birth 428 or 427 BC German Births per month Death 348 BC Relative population French size Categories: 420s BC births | 340s BC Summary deaths | Greek philosoph | Greek Adding Words to male wrestler | Austrian writer Regexes Answering Queries with Unix bornOn (infobox) Shell 428 Conclusion - 427 bornOn (category) ? 9/50 - 42#
Recommend
More recommend