Link Discovery Tutorial Introduction
2 likes1,118 views
The document provides an introduction to a tutorial on link discovery, focusing on basic ideas, principles, and challenges in the field. It discusses the importance of link discovery in enabling cross-ontology question answering, data integration, and complex information needs in linked data. The document also outlines definitions, challenges, and examples, while emphasizing the need for efficient methodologies due to the extensive amount of data involved.
![Introduction
Disclaimer
No pursuit of completeness [Nen+15]
Focus on
Basic ideas and principles
Principles
Evaluation
Open questions and challenges
Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 2 / 24](https://image.slidesharecdn.com/ldtutorialintro-161108150859/85/Link-Discovery-Tutorial-Introduction-2-320.jpg)
![Introduction
Disclaimer
No pursuit of completeness [Nen+15]
Focus on
Basic ideas and principles
Principles
Evaluation
Open questions and challenges
Guidelines
http://iswc2016ldtutorial.aksw.org/
Question? Just ask!
Comment? Go for it!
Be kind :)
Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 2 / 24](https://image.slidesharecdn.com/ldtutorialintro-161108150859/85/Link-Discovery-Tutorial-Introduction-3-320.jpg)
![Cross-Ontology QA
Example
Give me the name and description of all drugs that cure their side-effect. [SNA13]
1 Need information from
Drugbank (Drug description)
Sider (Side-effects)
DBpedia (Description)
2 Gathering information via SPARQL query
using links
Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 6 / 24](https://image.slidesharecdn.com/ldtutorialintro-161108150859/85/Link-Discovery-Tutorial-Introduction-7-320.jpg)
![Cross-Ontology QA
Example (DEQA)
Give me flats near kindergartens in Kobe. [Leh+12]
SELECT ?flat WHERE
{
?flat a deqa:Flat .
?flat deqa:near ?school .
?school a lgdo:School .
?school lgdo:city lgdo:Kobe .
}
Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 8 / 24](https://image.slidesharecdn.com/ldtutorialintro-161108150859/85/Link-Discovery-Tutorial-Introduction-9-320.jpg)
![Data Integration
Federated Queries on Patient Data [Kha+14]
Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 9 / 24](https://image.slidesharecdn.com/ldtutorialintro-161108150859/85/Link-Discovery-Tutorial-Introduction-10-320.jpg)
![Federated Queries
Example (FedBench CD2)
Return Barack Obama’s party membership and news pages. [Sal+15]
SELECT ?party ?page WHERE
{
dbr:Barack_Obama dbo:party ?party .
?x nytimes:topicPage ?page .
?x owl:sameAs dbr:Barack_Obama .
}
Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 10 / 24](https://image.slidesharecdn.com/ldtutorialintro-161108150859/85/Link-Discovery-Tutorial-Introduction-11-320.jpg)
![Definition
Most common: R = owl:sameAs
Also known as deduplication [Nen+15]
Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 12 / 24](https://image.slidesharecdn.com/ldtutorialintro-161108150859/85/Link-Discovery-Tutorial-Introduction-15-320.jpg)
![Why is it difficult?
1 Time complexity
Large number of triples (e.g.,
LinkedTCGA with 20.4 billion
triples [Sal+14])
Quadratic a-priori runtime
69 days for mapping cities from
DBpedia to Geonames
Solutions usually in-memory
(insufficient heap space)
Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 14 / 24](https://image.slidesharecdn.com/ldtutorialintro-161108150859/85/Link-Discovery-Tutorial-Introduction-18-320.jpg)
![Why is it difficult?
1 Time complexity
Large number of triples (e.g.,
LinkedTCGA with 20.4 billion
triples [Sal+14])
Quadratic a-priori runtime
69 days for mapping cities from
DBpedia to Geonames
Solutions usually in-memory
(insufficient heap space)
2 Accuracy
Combination of several attributes
required for high precision
Tedious discovery of most
adequate mapping
Dataset-dependent similarity
functions
Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 14 / 24](https://image.slidesharecdn.com/ldtutorialintro-161108150859/85/Link-Discovery-Tutorial-Introduction-19-320.jpg)
![Structure
1 Time complexity(≈ 60 min)
LIMES algorithm [NA11]
MultiBlock [IJB11]
HR3
[Ngo12]
AEGLE [GSN16]
Summary and Challenges
2 Accuracy(≈ 30 min)
RAVEN [Ngo+11]
EAGLE [NL12]
COALA [NLC13]
Summary and Challenges
3 Benchmarking (≈ 30 min)
Benchmarking [NGF16]
Synthetic Benchmarks [Sav+15]
Real Benchmarks [Mor+11]
Summary and Challenges
4 Hands-On Session (≈ 45 min)
Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 15 / 24](https://image.slidesharecdn.com/ldtutorialintro-161108150859/85/Link-Discovery-Tutorial-Introduction-20-320.jpg)
More Related Content
Similar to Link Discovery Tutorial Introduction (20)
More from Holistic Benchmarking of Big Linked Data (20)
Link Discovery Tutorial Introduction
- 1. Link Discovery Tutorial Introduction Axel-Cyrille Ngonga Ngomo(1) , Irini Fundulaki(2) , Mohamed Ahmed Sherif(1) (1) Institute for Applied Informatics, Germany (2) FORTH, Greece October 18th, 2016 Kobe, Japan Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 1 / 24
- 2. Introduction Disclaimer No pursuit of completeness [Nen+15] Focus on Basic ideas and principles Principles Evaluation Open questions and challenges Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 2 / 24
- 3. Introduction Disclaimer No pursuit of completeness [Nen+15] Focus on Basic ideas and principles Principles Evaluation Open questions and challenges Guidelines http://iswc2016ldtutorial.aksw.org/ Question? Just ask! Comment? Go for it! Be kind :) Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 2 / 24
- 4. Let’s Go! Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 3 / 24
- 5. Linked Data Principles Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 4 / 24
- 6. Why Link Discovery? 1 Linked Open Data Cloud 130+ billion triples ≈ 0.5 billion links Mostly owl:sameAs 2 Decentralized dataset creation 3 Complex information needs ⇒ Need to consume data across knowledge bases 4 Links are central for Cross-ontology QA Data Integration Reasoning Federated Queries ... Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 5 / 24
- 7. Cross-Ontology QA Example Give me the name and description of all drugs that cure their side-effect. [SNA13] 1 Need information from Drugbank (Drug description) Sider (Side-effects) DBpedia (Description) 2 Gathering information via SPARQL query using links Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 6 / 24
- 8. Cross-Ontology QA Example Give me the name and description of all drugs that cure their side-effect. SELECT ?drug ?name ?desc WHERE { ?drug a drugbank:Drug . ?drug rdfs:label ?name . ?drug drugbank:cures ?disease . ?drug owl:sameAs ?drug2 . ?drug owl:sameAs ?drug3 . ?drug2 sider:hasSideEffect ?effect . ?effect owl:sameAs ?disease . ?drug3 dbo:hasWikiPage ?desc . } Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 7 / 24
- 9. Cross-Ontology QA Example (DEQA) Give me flats near kindergartens in Kobe. [Leh+12] SELECT ?flat WHERE { ?flat a deqa:Flat . ?flat deqa:near ?school . ?school a lgdo:School . ?school lgdo:city lgdo:Kobe . } Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 8 / 24
- 10. Data Integration Federated Queries on Patient Data [Kha+14] Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 9 / 24
- 11. Federated Queries Example (FedBench CD2) Return Barack Obama’s party membership and news pages. [Sal+15] SELECT ?party ?page WHERE { dbr:Barack_Obama dbo:party ?party . ?x nytimes:topicPage ?page . ?x owl:sameAs dbr:Barack_Obama . } Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 10 / 24
- 12. Definition Definition (Link Discovery, informal) Given two knowledge bases S and T, find links of type R between S and T Here, declarative link discovery Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 11 / 24
- 13. Definition Definition (Link Discovery, informal) Given two knowledge bases S and T, find links of type R between S and T Here, declarative link discovery Definition (Declarative Link Discovery, formal, similarities) Given sets S and T of resources and relation R Find M = {(s, t) ∈ S × T : R(s, t)} Common approach: Find M = {(s, t) ∈ S × T : σ(s, t) ≥ θ} Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 11 / 24
- 14. Definition Definition (Link Discovery, informal) Given two knowledge bases S and T, find links of type R between S and T Here, declarative link discovery Definition (Declarative Link Discovery, formal, similarities) Given sets S and T of resources and relation R Find M = {(s, t) ∈ S × T : R(s, t)} Common approach: Find M = {(s, t) ∈ S × T : σ(s, t) ≥ θ} Definition (Declarative Link Discovery, formal, distances) Given sets S and T of resources and relation R Find M = {(s, t) ∈ S × T : R(s, t)} Common approach: Find M = {(s, t) ∈ S × T : δ(s, t) ≤ τ} Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 11 / 24
- 15. Definition Most common: R = owl:sameAs Also known as deduplication [Nen+15] Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 12 / 24
- 16. Definition Goal: Address all possible relations R Declarative Link Discovery: Similarity/distance defined using property values (incl. property chains) Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 13 / 24
- 17. Definition Goal: Address all possible relations R Declarative Link Discovery: Similarity/distance defined using property values (incl. property chains) Example: R = :sameModel :s770fm rdfs:label "S770FM"@en :s770fm rdf:type :SABER :s770fm :model :770 :s770fm :top :FlamedMaple :s770fm :producer :Ibanez :s770fm rdfs:label "S770BEM"@en :s770fm rdf:type :SABER :s770fm :model :770 :s770fm :top :BirdEyeMaple :s770fm :producer :Ibanez Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 13 / 24
- 18. Why is it difficult? 1 Time complexity Large number of triples (e.g., LinkedTCGA with 20.4 billion triples [Sal+14]) Quadratic a-priori runtime 69 days for mapping cities from DBpedia to Geonames Solutions usually in-memory (insufficient heap space) Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 14 / 24
- 19. Why is it difficult? 1 Time complexity Large number of triples (e.g., LinkedTCGA with 20.4 billion triples [Sal+14]) Quadratic a-priori runtime 69 days for mapping cities from DBpedia to Geonames Solutions usually in-memory (insufficient heap space) 2 Accuracy Combination of several attributes required for high precision Tedious discovery of most adequate mapping Dataset-dependent similarity functions Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 14 / 24
- 20. Structure 1 Time complexity(≈ 60 min) LIMES algorithm [NA11] MultiBlock [IJB11] HR3 [Ngo12] AEGLE [GSN16] Summary and Challenges 2 Accuracy(≈ 30 min) RAVEN [Ngo+11] EAGLE [NL12] COALA [NLC13] Summary and Challenges 3 Benchmarking (≈ 30 min) Benchmarking [NGF16] Synthetic Benchmarks [Sav+15] Real Benchmarks [Mor+11] Summary and Challenges 4 Hands-On Session (≈ 45 min) Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 15 / 24
- 21. That’s all Folks! Axel Ngonga AKSW Research Group Institute for Applied Informatics [email protected] Irini Fundulaki ICS FORTH [email protected] Mohamed Ahmed Sherif AKSW Research Group Institute for Applied Informatics [email protected] Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 16 / 24
- 22. Acknowledgment This work was supported by grants from the EU H2020 Framework Programme provided for the project HOBBIT (GA no. 688227). Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 17 / 24
- 23. References I Kleanthi Georgala, Mohamed Ahmed Sherif, and Axel-Cyrille Ngonga Ngomo. “An Efficient Approach for the Generation of Allen Relations”. In: ECAI 2016 - 22nd European Conference on Artificial Intelligence, 29 August-2 September 2016, The Hague, The Netherlands - Including Prestigious Applications of Artificial Intelligence (PAIS 2016). Ed. by Gal A. Kaminka et al. Vol. 285. Frontiers in Artificial Intelligence and Applications. IOS Press, 2016, pp. 948–956. isbn: 978-1-61499-671-2. doi: 10.3233/978-1-61499-672-9-948. url: http://dx.doi.org/10.3233/978-1-61499-672-9-948. Robert Isele, Anja Jentzsch, and Christian Bizer. “Efficient Multidimensional Blocking for Link Discovery without losing Recall”. In: Proceedings of the 14th International Workshop on the Web and Databases 2011, WebDB 2011, Athens, Greece, June 12, 2011. Ed. by Amélie Marian and Vasilis Vassalos. 2011. url: http://webdb2011.rutgers.edu/papers/Paper%2039/silk.pdf. Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 18 / 24
- 24. References II Yasar Khan et al. “SAFE: Policy Aware SPARQL Query Federation Over RDF Data Cubes”. In: Proceedings of the 7th International Workshop on Semantic Web Applications and Tools for Life Sciences, Berlin, Germany, December 9-11, 2014. Ed. by Adrian Paschke et al. Vol. 1320. CEUR Workshop Proceedings. CEUR-WS.org, 2014. url: http://ceur-ws.org/Vol-1320/Preface_SWAT4LS2014.pdf. Jens Lehmann et al. “deqa: Deep Web Extraction for Question Answering”. In: The Semantic Web - ISWC 2012 - 11th International Semantic Web Conference, Boston, MA, USA, November 11-15, 2012, Proceedings, Part II. Ed. by Philippe Cudré-Mauroux et al. Vol. 7650. Lecture Notes in Computer Science. Springer, 2012, pp. 131–147. isbn: 978-3-642-35172-3. doi: 10.1007/978-3-642-35173-0_9. url: http://dx.doi.org/10.1007/978-3-642-35173-0_9. Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 19 / 24
- 25. References III Mohamed Morsey et al. “DBpedia SPARQL Benchmark - Performance Assessment with Real Queries on Real Data”. In: The Semantic Web - ISWC 2011 - 10th International Semantic Web Conference, Bonn, Germany, October 23-27, 2011, Proceedings, Part I. Ed. by Lora Aroyo et al. Vol. 7031. Lecture Notes in Computer Science. Springer, 2011, pp. 454–469. isbn: 978-3-642-25072-9. doi: 10.1007/978-3-642-25073-6_29. url: http://dx.doi.org/10.1007/978-3-642-25073-6_29. Axel-Cyrille Ngonga Ngomo and Sören Auer. “LIMES - A Time-Efficient Approach for Large-Scale Link Discovery on the Web of Data”. In: IJCAI 2011, Proceedings of the 22nd International Joint Conference on Artificial Intelligence, Barcelona, Catalonia, Spain, July 16-22, 2011. Ed. by Toby Walsh. IJCAI/AAAI, 2011, pp. 2312–2317. isbn: 978-1-57735-516-8. doi: 10.5591/978-1-57735-516-8/IJCAI11-385. url: http: //dx.doi.org/10.5591/978-1-57735-516-8/IJCAI11-385. Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 20 / 24
- 26. References IV Markus Nentwig et al. “A survey of current Link Discovery frameworks”. In: Semantic Web Preprint (2015), pp. 1–18. Axel-Cyrille Ngonga Ngomo, Alejandra Garcıa-Rojas, and Irini Fundulaki. “HOBBIT: Holistic Benchmarking of Big Linked Data”. In: ERCIM News 2016.105 (2016). url: http://ercim-news.ercim.eu/en105/r-i/hobbit-holistic- benchmarking-of-big-linked-data. Axel-Cyrille Ngonga Ngomo et al. “RAVEN - active learning of link specifications”. In: Proceedings of the 6th International Workshop on Ontology Matching, Bonn, Germany, October 24, 2011. Ed. by Pavel Shvaiko et al. Vol. 814. CEUR Workshop Proceedings. CEUR-WS.org, 2011. url: http://ceur-ws.org/Vol-814/om2011_Tpaper3.pdf. Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 21 / 24
- 27. References V Axel-Cyrille Ngonga Ngomo. “Link Discovery with Guaranteed Reduction Ratio in Affine Spaces with Minkowski Measures”. In: The Semantic Web - ISWC 2012 - 11th International Semantic Web Conference, Boston, MA, USA, November 11-15, 2012, Proceedings, Part I. Ed. by Philippe Cudré-Mauroux et al. Vol. 7649. Lecture Notes in Computer Science. Springer, 2012, pp. 378–393. isbn: 978-3-642-35175-4. doi: 10.1007/978-3-642-35176-1_24. url: http://dx.doi.org/10.1007/978-3-642-35176-1_24. Axel-Cyrille Ngonga Ngomo and Klaus Lyko. “EAGLE: Efficient Active Learning of Link Specifications Using Genetic Programming”. In: The Semantic Web: Research and Applications - 9th Extended Semantic Web Conference, ESWC 2012, Heraklion, Crete, Greece, May 27-31, 2012. Proceedings. Ed. by Elena Simperl et al. Vol. 7295. Lecture Notes in Computer Science. Springer, 2012, pp. 149–163. isbn: 978-3-642-30283-1. doi: 10.1007/978-3-642-30284-8_17. url: http://dx.doi.org/10.1007/978-3-642-30284-8_17. Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 22 / 24
- 28. References VI Axel-Cyrille Ngonga Ngomo, Klaus Lyko, and Victor Christen. “COALA - Correlation-Aware Active Learning of Link Specifications”. In: The Semantic Web: Semantics and Big Data, 10th International Conference, ESWC 2013, Montpellier, France, May 26-30, 2013. Proceedings. Ed. by Philipp Cimiano et al. Vol. 7882. Lecture Notes in Computer Science. Springer, 2013, pp. 442–456. isbn: 978-3-642-38287-1. doi: 10.1007/978-3-642-38288-8_30. url: http://dx.doi.org/10.1007/978-3-642-38288-8_30. Muhammad Saleem et al. “TopFed: TCGA Tailored Federated Query Processing and Linking to LOD”. In: J. Biomedical Semantics 5 (2014), p. 47. doi: 10.1186/2041-1480-5-47. url: http://dx.doi.org/10.1186/2041-1480-5-47. Muhammad Saleem et al. “A fine-grained evaluation of SPARQL endpoint federation systems”. In: Semantic Web 7.5 (2015), pp. 493–518. doi: 10.3233/SW-150186. url: http://dx.doi.org/10.3233/SW-150186. Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 23 / 24
- 29. References VII Tzanina Saveta et al. “LANCE: Piercing to the Heart of Instance Matching Tools”. In: The Semantic Web - ISWC 2015 - 14th International Semantic Web Conference, Bethlehem, PA, USA, October 11-15, 2015, Proceedings, Part I. Ed. by Marcelo Arenas et al. Vol. 9366. Lecture Notes in Computer Science. Springer, 2015, pp. 375–391. isbn: 978-3-319-25006-9. doi: 10.1007/978-3-319-25007-6_22. url: http://dx.doi.org/10.1007/978-3-319-25007-6_22. Saeedeh Shekarpour, Axel-Cyrille Ngonga Ngomo, and Sören Auer. “Question answering on interlinked data”. In: 22nd International World Wide Web Conference, WWW ’13, Rio de Janeiro, Brazil, May 13-17, 2013. Ed. by Daniel Schwabe et al. International World Wide Web Conferences Steering Committee / ACM, 2013, pp. 1145–1156. isbn: 978-1-4503-2035-1. url: http://dl.acm.org/citation.cfm?id=2488488. Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial:Intro October 17, 2016 24 / 24