Link Discovery Tutorial Part V: Hands-On
0 likes827 views
This document provides an overview of hands-on tasks for a link discovery tutorial using the Limes framework. It describes a test dataset, and three tasks: 1) executing a provided Limes configuration to detect duplicate authors, 2) creating a configuration to find similar publications based on keywords, and 3) using the Limes GUI.
Link Discovery Tutorial Part V: Hands-On
- 1. Link Discovery Tutorial Part V: Hands-On 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: Hands-On October 17, 2016 1 / 19
- 2. Table of Contents 1 Test Dataset 2 Task I: Execute given Limes Configuration 3 Task II: Create your first Limes Configuration 4 Task III: Use Limes GUI Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial: Hands-On October 17, 2016 2 / 19
- 3. Table of Contents 1 Test Dataset 2 Task I: Execute given Limes Configuration 3 Task II: Create your first Limes Configuration 4 Task III: Use Limes GUI Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial: Hands-On October 17, 2016 3 / 19
- 4. Test Dataset Semantic Web Dog Food Corpus Data exposed: Metadata (papers, presentations, people) for several semantic web related conferences and workshops, including the most recent ISWC, ESWC and WWW events. Dumps: http://data.semanticweb.org/dumps Endpoint: http://data.semanticweb.org/sparql DataHub: https://datahub.io/dataset/semantic-web-dog-food Download: http://iswc2016ldtutorial.aksw.org/tutorial-material/ Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial: Hands-On October 17, 2016 4 / 19
- 5. Table of Contents 1 Test Dataset 2 Task I: Execute given Limes Configuration 3 Task II: Create your first Limes Configuration 4 Task III: Use Limes GUI Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial: Hands-On October 17, 2016 5 / 19
- 6. Task I: Deduplication Execute the given Configuration File Decentralized nature of LOD Data contain duplicates How to efficiently detect similar resources? Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial: Hands-On October 17, 2016 6 / 19
- 7. Task I: Deduplication Execute the given Configuration File Decentralized nature of LOD Data contain duplicates How to efficiently detect similar resources? Task I Find duplicate authors in Semantic Web Dog Food Dataset Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial: Hands-On October 17, 2016 6 / 19
- 8. Limes Configuration File PREFIXes <PREFIX > <NAMESPACE >http :// www.w3.org /2000/01/ rdf -schema#</NAMESPACE > <LABEL >rdfs </LABEL > </PREFIX > Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial: Hands-On October 17, 2016 7 / 19
- 9. Limes Configuration File SOURCE and TARGET datasets <SOURCE > <ID>SDF1 </ID> <ENDPOINT >semanticDogFood .nt</ENDPOINT > <VAR >?x</VAR> <PAGESIZE >-1</PAGESIZE > <RESTRICTION >?x a foaf:Person </ RESTRICTION > <PROPERTY >rdfs:label </PROPERTY > <TYPE >NT</TYPE > </SOURCE > Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial: Hands-On October 17, 2016 8 / 19
- 10. Limes Configuration File METRIC <METRIC >Levenshtein(x.rdfs:label , y.rdfs:label)</METRIC > Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial: Hands-On October 17, 2016 9 / 19
- 11. Limes Configuration File ACCEPTANCE and REVIEW <ACCEPTANCE > <THRESHOLD >0.9</THRESHOLD > <FILE >similarAuthor .nt</FILE > <RELATION >ov:similarTo </RELATION > </ACCEPTANCE > <REVIEW > <THRESHOLD >0.5 </THRESHOLD > <FILE >similarAuthor_review .nt</FILE > <RELATION >owl:sameAs </RELATION > </REVIEW > Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial: Hands-On October 17, 2016 10 / 19
- 12. Limes Configuration File OUTPUT format <OUTPUT >TTL</OUTPUT > Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial: Hands-On October 17, 2016 11 / 19
- 13. Task I Run Limes Run Limes java -jar limes-core-1.0.0.jar task1.xml Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial: Hands-On October 17, 2016 12 / 19
- 14. Task I Using Machine Learning <MLALGORITHM > <NAME >wombat simple </NAME > <TYPE >unsupervised </TYPE > </ MLALGORITHM > Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial: Hands-On October 17, 2016 13 / 19
- 15. Table of Contents 1 Test Dataset 2 Task I: Execute given Limes Configuration 3 Task II: Create your first Limes Configuration 4 Task III: Use Limes GUI Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial: Hands-On October 17, 2016 14 / 19
- 16. Task II Find similar publications 1 Find publications 2 with similar keyword, but do not link any publication to itself Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial: Hands-On October 17, 2016 15 / 19
- 17. Task II Find similar publications 1 Find publications 2 with similar keyword, but do not link any publication to itself 1. Find publications ?x a swrc:InProceedings ?y a swrc:InProceedings Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial: Hands-On October 17, 2016 15 / 19
- 18. Task II Find similar publications 1 Find publications 2 with similar keyword, but do not link any publication to itself 1. Find publications ?x a swrc:InProceedings ?y a swrc:InProceedings 2. with similar keyword, but do not link any publication to itself MINUS(jaccard(x.swrc:listKeyword , y.swrc:listKeyword)|0.3, ExactMatch(x.swrc:listKeyword , y.swrc:listKeyword)|1.0) Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial: Hands-On October 17, 2016 15 / 19
- 19. Table of Contents 1 Test Dataset 2 Task I: Execute given Limes Configuration 3 Task II: Create your first Limes Configuration 4 Task III: Use Limes GUI Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial: Hands-On October 17, 2016 16 / 19
- 20. Task III Use Limes GUI Run Limes GUI java -jar limes-core-1.0.0.jar -g Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial: Hands-On October 17, 2016 17 / 19
- 21. 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: Hands-On October 17, 2016 18 / 19
- 22. References I Ngonga Ngomo et al. (InfAI & FORTH) LD Tutorial: Hands-On October 17, 2016 19 / 19