306
M. Noura et al.
Contributions are as follows: Our innovative Knowledge Extraction for the Automotive Sector (KEAS) methodology understands the “common sense knowledge”
required to build smart vehicle applications which provides:
1. C1: A set of keyphrase synonyms for the smart vehicle domain to find domainspecific knowledge in past or current projects that published their results within
scientific publications,
2. C2: Synonyms are used to build a corpus of scientific publications to train the
k-means machine learning algorithm,
3. C3: A dataset of smart car ontologies is built and analyzed by the k-means
algorithm to cluster knowledge, and,
4. C4: The extraction of the most common knowledge for the automotive sector. We
refined a previous methodology [9] that we applied to the smart vehicle domain
in this book chapter.
Structure of the Paper: Sect. 3 introduces the related work. Section 4 explains
our Knowledge Extraction for the Automotive Sector (KEAS) methodology to find
the relevant knowledge already implemented within ontologies. Section 5 evaluates
our proposed approach. Section 6 concludes the paper and provides future work.
3 Background and Related Work
Toyota Motor Europe (TME)
11 uses auto.schema.org in their web site to describe
cars to sell. For instance, 7000 URLs including the type “Car” from the TME web
site have been encoded and indexed by Google.
auto.schema.org
12 defines 4 types (BusOrCoach, CarUsageType, Motorcycle,
MotorizedBicycle), 20 properties (accelerationTime, acrissCode, bodyType,
emissionsCO2, engineDisplacement, enginePower, engineType, fuelCapacity,
meetsEmissionStandard, modelDate, payload, roofLoad, seatingCapacity, speed,
tongueWeight, torque, trailerWeight, vehicleSpecialUsage, weightTotal, wheelbase)
and 3 enumeration values (DrivingSchoolVehicleUsage, RentalVehicleUsage, TaxiVehicleUsage) (in December 2018). It clearly shows the the knowledge could be
extended.
OpenSensingCity
13 references 12 ontology URLs relevant to mobility: Transport, travel domain, transportation networks, transport disruption, soft mobility, PASSIM, location concept for travel support system, route, ASK-IT, road, transit.
SAREF4AUTO is being specified and supported by the ETSI standard; the ontology code and specification cannot be found yet, only those slides can be investigated
[10] at the time of this writing.
Conclusion: Ontology-based projects are introduced in Table 1 when ontologies
are publicly available, that we analyze thanks to the KEAS methodology in Sect. 4.2.
11 http://bit.ly/2Y3A1xL.
12 https://auto.schema.org/.
13 http://ci.emse.fr/opensensingcity/ns/result/domain/transportation/.
M. Noura et al.
Contributions are as follows: Our innovative Knowledge Extraction for the Automotive Sector (KEAS) methodology understands the “common sense knowledge”
required to build smart vehicle applications which provides:
1. C1: A set of keyphrase synonyms for the smart vehicle domain to find domainspecific knowledge in past or current projects that published their results within
scientific publications,
2. C2: Synonyms are used to build a corpus of scientific publications to train the
k-means machine learning algorithm,
3. C3: A dataset of smart car ontologies is built and analyzed by the k-means
algorithm to cluster knowledge, and,
4. C4: The extraction of the most common knowledge for the automotive sector. We
refined a previous methodology [9] that we applied to the smart vehicle domain
in this book chapter.
Structure of the Paper: Sect. 3 introduces the related work. Section 4 explains
our Knowledge Extraction for the Automotive Sector (KEAS) methodology to find
the relevant knowledge already implemented within ontologies. Section 5 evaluates
our proposed approach. Section 6 concludes the paper and provides future work.
3 Background and Related Work
Toyota Motor Europe (TME)
11 uses auto.schema.org in their web site to describe
cars to sell. For instance, 7000 URLs including the type “Car” from the TME web
site have been encoded and indexed by Google.
auto.schema.org
12 defines 4 types (BusOrCoach, CarUsageType, Motorcycle,
MotorizedBicycle), 20 properties (accelerationTime, acrissCode, bodyType,
emissionsCO2, engineDisplacement, enginePower, engineType, fuelCapacity,
meetsEmissionStandard, modelDate, payload, roofLoad, seatingCapacity, speed,
tongueWeight, torque, trailerWeight, vehicleSpecialUsage, weightTotal, wheelbase)
and 3 enumeration values (DrivingSchoolVehicleUsage, RentalVehicleUsage, TaxiVehicleUsage) (in December 2018). It clearly shows the the knowledge could be
extended.
OpenSensingCity
13 references 12 ontology URLs relevant to mobility: Transport, travel domain, transportation networks, transport disruption, soft mobility, PASSIM, location concept for travel support system, route, ASK-IT, road, transit.
SAREF4AUTO is being specified and supported by the ETSI standard; the ontology code and specification cannot be found yet, only those slides can be investigated
[10] at the time of this writing.
Conclusion: Ontology-based projects are introduced in Table 1 when ontologies
are publicly available, that we analyze thanks to the KEAS methodology in Sect. 4.2.
11 http://bit.ly/2Y3A1xL.
12 https://auto.schema.org/.
13 http://ci.emse.fr/opensensingcity/ns/result/domain/transportation/.