How to Understand Better “Smart Vehicle”? Knowledge Extraction …
305
ontology.
6 auto.schema.org
7 defines 4 types, 20 properties and 3 enumeration values
(in December 2018) which clearly shows that the knowledge could be extended.
Volvo is investigating the integration of semantic web technologies (RDF, Linked
Data, ontologies) for automomous cars.
8
Acquiring knowledge about automotive (e.g., technological survey, reading scientific publications and staying updating with the latest progresses) is a time-consuming
approach. The survey about transportation ontologies [7], published in 2018, can be
easily enriched with numerous ontologies that we collected within the LOV4IoT
ontology catalog for IoT and transport
9 that we designed. The survey [7] compares
11 ontologies according to 7 criteria: (1) Precision (relationship diversity, axiom
complexity), (2) Evaluation, (3) Knowledge management services, (4) Generality,
(5) Granularity, (6) Competence, and (7) Span.
We designed the “semantic-based IoT smart vehicle” LOV4IoT dataset thats collects common sense knowledge for the automative sector. We classified 42 projects
between 2005 and 2019 since they claim that the knowledge is already encoded with
knowledge representation languages (e.g., RDF, RDFS, and OWL) and supported by
the World Wide Web Consortium (W3C). However, only a subset of those projects
share their expertise by publishing their ontologies online. For this reason, at the
current time or writing, our dataset comprises only 16 processable ontologies.
Motivation are as follows:
• M1: Why cannot we find the entire Ph.D. thesis, entitled “Using Ontologies and
Intelligent Systems for Traffic Accident Assistance in Vehicular Environments”
[8] published in 2014 relevant for smart car on the first page of Google results? It
is provided on the third page on Google
10 whereas years of research and expertise
are explained in the thesis.
• M2: How to find more knowledge than Google for a specific domain (e.g, smart
vehicle)?
• M3: Why does the Google Knowledge Graph cannot provide results to handle the
synonyms used for the automotive domain (e.g., smart car, smart vehicle, smart
mobility)?
Research questions are as follows:
• RQ1: How to automatically analyze structured knowledge (e.g., ontologies) from
existing projects? We found that numerous projects designed ontologies that are
also explained within scientific publications can be analyzed.
• RQ2: What are the most used entities (e.g, concepts, instances) within those ontologies? Statistical methods can help to achieve this task.
6 http://automotive.eurecom.fr/vdc.
7 https://auto.schema.org/.
8 https://twitter.com/olafhartig/status/1121539105924550661.
9 http://lov4iot.appspot.com/?p=lov4iot-transport.
10 “Smart car ontology” search on Google, December 2018.
305
ontology.
6 auto.schema.org
7 defines 4 types, 20 properties and 3 enumeration values
(in December 2018) which clearly shows that the knowledge could be extended.
Volvo is investigating the integration of semantic web technologies (RDF, Linked
Data, ontologies) for automomous cars.
8
Acquiring knowledge about automotive (e.g., technological survey, reading scientific publications and staying updating with the latest progresses) is a time-consuming
approach. The survey about transportation ontologies [7], published in 2018, can be
easily enriched with numerous ontologies that we collected within the LOV4IoT
ontology catalog for IoT and transport
9 that we designed. The survey [7] compares
11 ontologies according to 7 criteria: (1) Precision (relationship diversity, axiom
complexity), (2) Evaluation, (3) Knowledge management services, (4) Generality,
(5) Granularity, (6) Competence, and (7) Span.
We designed the “semantic-based IoT smart vehicle” LOV4IoT dataset thats collects common sense knowledge for the automative sector. We classified 42 projects
between 2005 and 2019 since they claim that the knowledge is already encoded with
knowledge representation languages (e.g., RDF, RDFS, and OWL) and supported by
the World Wide Web Consortium (W3C). However, only a subset of those projects
share their expertise by publishing their ontologies online. For this reason, at the
current time or writing, our dataset comprises only 16 processable ontologies.
Motivation are as follows:
• M1: Why cannot we find the entire Ph.D. thesis, entitled “Using Ontologies and
Intelligent Systems for Traffic Accident Assistance in Vehicular Environments”
[8] published in 2014 relevant for smart car on the first page of Google results? It
is provided on the third page on Google
10 whereas years of research and expertise
are explained in the thesis.
• M2: How to find more knowledge than Google for a specific domain (e.g, smart
vehicle)?
• M3: Why does the Google Knowledge Graph cannot provide results to handle the
synonyms used for the automotive domain (e.g., smart car, smart vehicle, smart
mobility)?
Research questions are as follows:
• RQ1: How to automatically analyze structured knowledge (e.g., ontologies) from
existing projects? We found that numerous projects designed ontologies that are
also explained within scientific publications can be analyzed.
• RQ2: What are the most used entities (e.g, concepts, instances) within those ontologies? Statistical methods can help to achieve this task.
6 http://automotive.eurecom.fr/vdc.
7 https://auto.schema.org/.
8 https://twitter.com/olafhartig/status/1121539105924550661.
9 http://lov4iot.appspot.com/?p=lov4iot-transport.
10 “Smart car ontology” search on Google, December 2018.