Catch the Wave: SAP Event-Driven and Data Streaming for the Intelligence Ente...
LogMap in Practice OM 2019
1. Using LogMap in Practice
Ernesto Jiménez-Ruiz, City, University of London, UK
26/10/2019 International Semantic Web Conference, Auckland, NZ
Using LogMap in Practice
1
2. LogMap in a Nutshell
– performs a simple matching but efficient (based on sophisticated
lexical and structural indexes),
– can efficiently match semantically rich ontologies containing
hundreds of thousands of classes (FMA, NCI and SNOMED),
– incorporates reasoning and repair capabilities, and
– is one of the top systems in the OAEI evaluation campaign.
– Core implementation in 2011-2012.
E. Jiménez-Ruiz et al. Large-scale Interactive Ontology Matching: Algorithms and Implementation. ECAI 2012
E. Jiménez-Ruiz et al. LogMap: Logic-based and Scalable Ontology Matching. ISWC 2011
26/10/2019 International Semantic Web Conference, Auckland, NZ
Using LogMap in Practice
2
3. Outline
– Repair of Mapping Repositories: UMLS and BioPortal
– Lung-Cancer Assistant
– Ontology-based Data Access: Oil & Gas
– Ecotoxicology
– Interoperability in the Lab
26/10/2019 International Semantic Web Conference, Auckland, NZ
Using LogMap in Practice
3
4. Repair of Mapping Repositories
26/10/2019 International Semantic Web Conference, Auckland, NZ
Using LogMap in Practice
4
5. Mapping Repositories
– UMLS (Unified Medical Language System) Metathesaurus
– Integrates more than one hundred thesauri and ontologies
– Contains more than 6 million entities
– BioPortal
– Contains more than 500 ontologies
– Represent a network of ontologies
– More than 10 million mappings are available
– Also includes user-submitted alignments
26/10/2019 International Semantic Web Conference, Auckland, NZ
Using LogMap in Practice
5
6. Mapping Repositories: limitations
Using OWL 2 semantics of the ontologies and mappings. . .
– may lead to undesired cross-references.
26/10/2019 International Semantic Web Conference, Auckland, NZ
Using LogMap in Practice
6
7. Mapping Repositories: limitations
Using OWL 2 semantics of the ontologies and mappings. . .
– may lead to logical errors.
26/10/2019 International Semantic Web Conference, Auckland, NZ
Using LogMap in Practice
7
8. Logic-based Assessment of UMLS
– Assessment of the integration of FMA, NCI and SNOMED CT
ontologies within UMLS
UMLS alignments
Ontologies # Logical errors
FMA-NCI 3,024 655
FMA-SNOMED 9,072 6,179
SNOMED-NCI 19,622 20,944
E. Jiménez-Ruiz, et al. Logic-based assessment of the compatibility of UMLS ontology sources. J. Biomedical Semantics,
2011
E. Jiménez-Ruiz, Bernardo Cuenca Grau. LogMap: Logic-based and Scalable Ontology Matching. ISWC 2011.
26/10/2019 International Semantic Web Conference, Auckland, NZ
Using LogMap in Practice
8
9. Logic-based Assessment of BioPortal
BioPortal alignments
Ontologies # Logical errors
BDO-NCIT 1,636 34,341
CCONT-NCIT 2,097 50,304
EFO-NCIT 2,507 60,347
EP-FMA 78,489 210
EP-NCIT 2,465 14,687
MA-FMA 961 850
OMIM-NCIT 5,178 70,172
SDO-EP 135 44
UBERON-FMA 1,932 4,753
ZFA-EFO 427 913
ZFA-UBERON 724 104
D. Faria, E. Jiménez-Ruiz, et al. Towards annotating potential incoherences in BioPortal mappings. ISWC 2014.
26/10/2019 International Semantic Web Conference, Auckland, NZ
Using LogMap in Practice
9
11. Lung Cancer Assistant (LCA)
– An ontology-based system which provides decision support for lung
cancer treatment
– LCA exploits the English Lung Cancer Dataset (LUCADA)
– LUCADA ontology represents the semantic layer of the LCA,
M. Berkan Sesen et al. Lung Cancer Assistant: a hybrid clinical decision support application for lung cancer care. Journal
of the Royal Society Interface. 2014.
26/10/2019 International Semantic Web Conference, Auckland, NZ
Using LogMap in Practice
11
12. Lung Cancer Assistant (LCA)
– An ontology-based system which provides decision support for lung
cancer treatment
– LCA exploits the English Lung Cancer Dataset (LUCADA)
– LUCADA ontology represents the semantic layer of the LCA,
– Required alignment with SNOMED CT
– to facilitate interoperability with NHS systems
– partially done without tool support!
M. Berkan Sesen et al. Lung Cancer Assistant: a hybrid clinical decision support application for lung cancer care. Journal
of the Royal Society Interface. 2014.
26/10/2019 International Semantic Web Conference, Auckland, NZ
Using LogMap in Practice
11
13. Lung Cancer Assistant (LCA)
– An ontology-based system which provides decision support for lung
cancer treatment
– LCA exploits the English Lung Cancer Dataset (LUCADA)
– LUCADA ontology represents the semantic layer of the LCA,
– Required alignment with SNOMED CT
– to facilitate interoperability with NHS systems
– partially done without tool support!
– We offered LogMap alignment system to
– identify the classes in SNOMED CT related to LUCADA
– extract a lung cancer-specific module of SNOMED CT
M. Berkan Sesen et al. Lung Cancer Assistant: a hybrid clinical decision support application for lung cancer care. Journal
of the Royal Society Interface. 2014.
26/10/2019 International Semantic Web Conference, Auckland, NZ
Using LogMap in Practice
11
14. Ontology-based Data Access: Oil & Gas
26/10/2019 International Semantic Web Conference, Auckland, NZ
Using LogMap in Practice
12
15. Ontology-based Data Access: Oil & Gas
EU Optique project
Preservation of entailments in DB-Ontology was required:
A. Solimando, E. Jiménez-Ruiz and G, Guerrini. Minimizing conservativity violations in ontology alignments: algorithms
and evaluation. Knowledge and Information Systems 2017
E. Kharlamov et al. Ontology Based Data Access in Statoil. Journal of Web Semantics 2017
26/10/2019 International Semantic Web Conference, Auckland, NZ
Using LogMap in Practice
13
16. Ecotoxicological Effect Prediction
– Knowledge Graph Construction
for Ecotoxicological Effect
Prediction
– Integration of several resources
relevant to species and
chemicals.
– ECOTOX contains data about
pairs Chemical-Species
E. B. Myklebust, E. Jimenez-Ruiz et al. Knowledge Graph Embedding for Ecotoxicological Effect Prediction. ISWC In-Use,
DC and Poster 2019
26/10/2019 International Semantic Web Conference, Auckland, NZ
Using LogMap in Practice
14
17. Interoperability in the Lab
26/10/2019 International Semantic Web Conference, Auckland, NZ
Using LogMap in Practice
15
18. Pistoia Alliance Ontologies Mapping project
– Not-for-profit alliance of life science companies, vendors, publishers,
and academics.
– Motivation: better integration, understanding and analysis of data
– Interest in ontology alignment in the:
– Disease and phenotype: OAEI’s phenotype track.
– Laboratory analytics domain: New OAEI track next year.
I. Harrow, E. Jimenez-Ruiz, et al. Matching Disease and Phenotype Ontologies in the Ontology Alignment Evaluation
Initiative. J. Biomedical Semantics 2018
I. Harrow, R. Balakrishnan, E. Jimenez-Ruiz et al. Ontology mapping for semantically enabled applications. Drug Discovery
Today, 2019
26/10/2019 International Semantic Web Conference, Auckland, NZ
Using LogMap in Practice
16
20. Future Extensions
– Extensions for Tabular Data to KG Matching
– Extension of the repair module
– DeepLogMap?:
– Integration with OWL2Vec to compute embeddings of OWL 2
ontologies
O. M. Holter, E. B. Myklebust, J. Chen and E. Jimenez-Ruiz. Embedding OWL Ontologies with OWL2Vec. ISWC Poster 2019.
26/10/2019 International Semantic Web Conference, Auckland, NZ
Using LogMap in Practice
18
21. Acknowledgements
– Prof. Bernardo Cuenca Grau and Prof. Ian Horrocks
– EPSRC LogMap, EU SEALS and EU Optique projects.
– The Alan Turing Institute
– SIRIUS Centre for Scalable Data Access
– LogMap contributors: Alessandro Solimando, Valerie Cross, Anton
Morant, Yujiao Zhou, Weiguo Xia, Xi Chen, Yuan Gong and Shuo
Zhang.
26/10/2019 International Semantic Web Conference, Auckland, NZ
Using LogMap in Practice
19
22. Questions?
– Ernesto Jiménez Ruiz
– ernesto.jimenez-ruiz@city.ac.uk
– ernesto.jimenez.ruiz@gmail.com
– https://www.city.ac.uk/people/academics/ernesto-jimenez-ruiz
26/10/2019 International Semantic Web Conference, Auckland, NZ
Using LogMap in Practice
20