Reusing and Unifying Background Knowledge for Internet of Things with LOV4IoT FiCloud 22-24 August 2016,Vienna, Austria Amelie Gyrard, Insight, Ireland Ghislain Atemezing, Mondeca, France Christian Bonnet, Eurecom, France Karima Boudaoud, University of Nice Sophia Antipolis France Martin Serrano, Insight, Ireland
20
Embed
Reusing and Unifying Background Knowledge for Internet of Things with LOV4IoT
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Reusing and Unifying Background Knowledge for Internet of Things with
LOV4IoT
FiCloud 22-24 August 2016,Vienna, Austria
Amelie Gyrard, Insight, Ireland Ghislain Atemezing, Mondeca, France Christian Bonnet, Eurecom, France Karima Boudaoud, University of Nice Sophia Antipolis France Martin Serrano, Insight, Ireland
Agenda
• Introduction & Motivation
• Contribution: LOV4IoT: Linked Open Vocabularies for Internet of Things LOV4IoT RDF dataset to make statistics Extracting domain knowledge
• Use Case & Evaluation
• Conclusion & Future work
2
Motivation: How to reuse Internet of Things applications?
Motivation: How to reuse domain knowledge already designed in previous IoT applications?
Classify Interoperability
Collect
How to exploit the domain knowledge already available on the Web and make it interoperable?
Basics: Semantic Web Technologies
• Domain knowledge already structured and designed
• Ontologies used to share and reuse the domain knowledge
Descriptions Tools
Feature
Pros
Cons
LOV - Ontology catalogue
- More than 469 ontologies referenced
- Ontologies designed by semantic web experts
- Not referenced if LOV recommendations are not followed
- Semi-automatic - IoT domain limited
DataHub - Dataset catalogue
- 9,195 datasets - Various format accepted
- IoT domain limited - No quality checked - Manually
READY4 SmartCities
- Ontology & dataset catalogue
- More than 50 projects referenced
- Manually
LOV4IoT (previous version)
- Domain knowledge
relevant for IoT
- More than 200 projects referenced
- Ontologies designed by domain experts
- Manually
Sindice, Watson, Swoogle
- Semantic Search engines
- Automatic tools - IoT domain limited - Project not referenced if
knowledge not available on the web.
Related Work
LOV4IoT: An extension of the LOV catalogue
7
• LOV4IoT: Linked Open Vocabularies for Internet of Things (LOV4IoT) • an extension of the Linked Open Vocabularies (LOV) catalogue o Numerous ontologies relevant for IoT were not referenced yet due
to a lack of unknown semantic web best practices
• LOV4IoT: a dataset referencing more than 300 ontology-based projects relevant for IoT o Ontologies, Datasets, Rules, Technologies, Sensors and
• To make statistics on the LOV4IoT dataset • To make LOV4IoT more automatic
LOV4IoT Web services: Automatically compute the number of projects per domain
Web service called: http://sensormeasurement.appspot.com/lov4iot/nbOntoDomain/?domain=BuildingAutomation
Display the result returned by the web service:
LOV4IoT Web services: Automatically compute the number of ontologies according to the semantic web best practices
Web service called: http://sensormeasurement.appspot.com/lov4iot/ontoStatus/?status=Online
Display the result returned by the web service:
LOV4IoT RDF dataset
Use our web service to automatically send email to encourage domain experts to share their domain knowledge
=> To replace by the author’s email
=> To replace by the title of the research article describing ontologies, datasets or rules relevant for IoT
LOV4IoT bot & Web Service
Extracting domain knowledge from LOV4IoT
• Extracting a dictionary to unify IoT data • Extracting IoT domains • Extracting rules to interpret data • Extracting knowledge from ontologies and datasets
LOV4IoT: Use Cases
LOV4IoT Use Case: Semantic Web of Things (SWoT) Generator
• Assisting developers in designing semantic-based IoT application by generating a template