Folie 1

1
YAGO A Core of Semantic Knowledge
Fabian M. Suchanek, Gjergji Kasneci, Gerhard
Weikum (Max-Planck Institute for Computer Science
Saarbrücken/Germany)
2
Overview

• ? Motivation
• ? The Yago ontology
• ? Content
• ? Model
• ? Conclusion

3
The Truth about Elvis

• Elvis is alive!

4
The Truth about Elvis

• Elvis is alive!
• He works as an astronaut in
• NASA's special security program

5
Usual solution
Which NASA astronaut was born when Elvis was born?
Yields only rubbish. Reasons
1. Google participates in the conspiracy
2. Google does not search knowledge, but Web sites
6
Solution An ontology
astronaut
is an
?
born
born
1935
7
Solution An ontology
entity
subclass
person
subclass
is a
astronaut
is a
?
born
born
1935
means
means
"Elvis Presley"
"The King"
8
Solution An ontology
entity
subclass
Classes
person
subclass
Relations
is a
astronaut
is a
?
born
born
Individuals
1935
means
means
"Elvis Presley"
"The King"
Words
9
Where do we get the ontology from?

• Previous approaches
• ? Assemble the ontology manually
• (WordNet, SUMO, GeneOntology)
• Problems Usually low coverage (MPI is in none
  of these)

? Extract the ontology from corpora (e.g. the
Web) (KnowItAll, Espresso, Snowball, LEILA)
Problem Usually low accuracy (50-92)
10
Where do we get the ontology from?

• YAGO approach
• Assemble the ontology from Wikipedia (gt good
  coverage)

Use the category system of Wikipedia (gt good
accuracy)
11
Exploiting the Wikipedia category system
Elvis Pr
born
1935
blah blah blub Elvis (don't read this! Better
listen to the talk!) laber fasel suelz.
Insbesondere, blub, texte zu, und so weiter blah
blah blub Elvis laber fasel suelz. Blub, aber
blah! Insbesondere, blub, texte zu, und so weiter
blah blah blub Elvis laber fasel suelz.
Insbesondere, blub, texte zu, und so weiter
Exploit relational categories
Categories
1935_births
12
Exploiting the Wikipedia category system
American_singer
Elvis Pr
is a
born
1935
blah blah blub Elvis (don't read this! Better
listen to the talk!) laber fasel suelz.
Insbesondere, blub, texte zu, und so weiter blah
blah blub Elvis laber fasel suelz. Blub, aber
blah! Insbesondere, blub, texte zu, und so weiter
blah blah blub Elvis laber fasel suelz.
Insbesondere, blub, texte zu, und so weiter
Exploit relational categories
Exploit conceptual categories
Categories
American_singers
13
Exploiting the Wikipedia category system
Disputed_article
American_singer
Elvis Pr
is a
is a
born
1935
blah blah blub Elvis (don't read this! Better
listen to the talk!) laber fasel suelz.
Insbesondere, blub, texte zu, und so weiter blah
blah blub Elvis laber fasel suelz. Blub, aber
blah! Insbesondere, blub, texte zu, und so weiter
blah blah blub Elvis laber fasel suelz.
Insbesondere, blub, texte zu, und so weiter
Exploit relational categories
Exploit conceptual categories
Categories
Avoid administrational categories
Disputed_articles
14
Exploiting the Wikipedia category system
Rock'n_Roll_Music
American_singer
Elvis Pr
is a
is a
born
1935
blah blah blub Elvis (don't read this! Better
listen to the talk!) laber fasel suelz.
Insbesondere, blub, texte zu, und so weiter blah
blah blub Elvis laber fasel suelz. Blub, aber
blah! Insbesondere, blub, texte zu, und so weiter
blah blah blub Elvis laber fasel suelz.
Insbesondere, blub, texte zu, und so weiter
Exploit relational categories
Exploit conceptual categories
Categories
Avoid administrational categories
Rock'n_Roll_Music
Avoid thematic categories
15
Thematic vs Conceptual Categories
American singers of German origin
Shallow linguistic noun phrase parsing
Premodifier Head Postmodifier
Heuristics If the head is a plural word, the
category is conceptual
16
The Upper Model
entity
?
person
American_singer
is a
born
1935
17
The Upper Model From Wikipedia?
Business
Social_group
?
People_by_occupation
American_singer
is a
born
1935
18
The Upper Model From WordNet?
Person3
Singer17
Singer1
...
?
American_singer
is a
born
1935
19
The Upper Model From WordNet?
Person3
Singer17
Singer1
...
!
American_singer
is a
born
1935
20
The YAGO ontology
Person3
subclass
Singer1
means
subclass
"singer"
American_singer
is a
born
1935
"Elvis Presley"
means
21
The YAGO ontology Accuracy
See TechReport for details on the evaluation.
22
The YAGO ontology Number of Facts
6,000,000
Ontologies should not be judged purely by the
number of facts! This is just an informational
overview.
2,000,000
30,000 60,000 200,000 300,000
KnowItAll SUMO WordNet OpenCyc Cyc
Yago
23
The Yago Model Why binary is not enough
singer
(Elvis, is_a, singer)
(But only from 1953 to 1977)
is a
(We know this from Wikipedia)
24
The Yago Model Why binary is not enough
singer
1 (Elvis, is_a, singer) 2 (1, time,
1953-1977) 3 (1, source, Wikipedia)
time
1953-1977
is a
source
Wikipedia
25
The Yago model formally

• A YAGO ontology over
• a set of relations R
• a set of common entities C
• a set of fact identifiers I
• is a function
• I ? (R?C?I) ? R ? (R?I?C)

1 (Elvis, is_a, singer) 2 (1, time,
1953-1977) 3 (1, source, Wikipedia)

• We can talk about
• facts (1, source, Wikipedia)
• additional arguments (1, time, 1953-1977)
• relations (time, hasRange, time_interval)

26
The Yago model Logical aspects
Axioms (x, is_a, y) (y, subclass, z) gt (x,
is_a, z) ...
person
subclass
singer
is a
is a
27
The Yago model Logical aspects
finite, unique
f1, f2, f3, f4, f5, f6, f7, f8, f9, f10
Axioms (x, is_a, y) (y, subclass, z) gt (x,
is_a, z) ...
derive facts
f1, f2, f3, f4, f5
Eliminate facts
f1, f2, f3
finite, unique
28
The Truth about Elvis
Which astronaut was born in the same year as
Elvis?
http//www.mpi-inf.mpg.de/suchanek/downloads/yago
/
Enter your Yago Query
"Elvis Presley" bornInYear year astro
bornInYear year astro isa astronaut
20 results
29
The Truth about Elvis
Which astronaut codenamed "Roger" was born in the
same year as Elvis?
http//www.mpi-inf.mpg.de/suchanek/downloads/yago
/
Enter your Yago Query
"Elvis Presley" bornInYear year astro
bornInYear year "Roger" givenNameOf
astro astro isa astronaut
astro Roger_Chaffee
30
Conclusions

• Yago bases on a logically clean model
• Yago has an accuracy of around 95
• Yago is 3 times larger than the largest
  competitor

? Elvis is alive
31
Reference
For all details, please refer to our technical
report "Yago A Core of Semantic
Knowledge" (Fabian M. Suchanek, Gjergji Kasneci,
Gerhard Weikum) available at http//www.mpii.mpg.d
e/suchanek BibTex _at_TECHREPORTyagotr,
AUTHOR Suchanek, Fabian and Kasneci, Gjergji
and Weikum, Gerhard, TITLE Yago A Core
of Semantic Knowledge, TYPE Research
Report, INSTITUTION Max-Planck-Institut
f\"ur Informatik, ADDRESS
Stuhlsatzenhausweg 85, 66123 Saarbr\"ucken,
Germany, NUMBER MPI-I-2006-5-006,
YEAR 2006