Towards Semantic Interoperability in WoT Ecosystems · WoT WoT Mappings Client SPARQL Resources...

21
Andrea Cimmino, María Poveda-Villalón, Raúl García- Castro Ontology Engineering Group Universidad Politécnica de Madrid, Spain Towards Semantic Interoperability in WoT Ecosystems [email protected] This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement no. 688467

Transcript of Towards Semantic Interoperability in WoT Ecosystems · WoT WoT Mappings Client SPARQL Resources...

Andrea Cimmino, María Poveda-Villalón, Raúl García-Castro

Ontology Engineering GroupUniversidad Politécnica de Madrid, Spain

Towards Semantic Interoperability in WoT Ecosystems

[email protected]

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement no. 688467

Set your title here in Slide Master Options

The Web of …

2

Set your title here in Slide Master Options

The Web of Heterogeneities

3

File storage

REST API

REST API

Database

Heterogeneus Access

Set your title here in Slide Master Options

The Web of Heterogeneities

4

File storage

REST API

REST API

Database

Heterogeneus Access Heterogeneus Formats

CSV

JSON

JSON

XML

Set your title here in Slide Master Options

The Web of Heterogeneities

5

{ "tmp" : "-2ºC","power" : "10Kw",”content": [

{ "type" : "apple"},{ "type" : "apple"},{ "type" : "orange"},

...] }

{ ”cold" : 268.15,“Kw” : 8,“Kw/€” : 50“objects” : 10

}

<caption><value> 3 </value></caption>

time,value00:20, 12.2900:25, 12.3000:30, 12.28

File storage

REST API

REST API

Database

Heterogeneus Access Heterogeneus Formats Heterogeneus Models

CSV

JSON

JSON

XML

Set your title here in Slide Master Options

Index

6

§ Bringing Semantic Interoperability§ The WoT-Mappings§ Semantic Interoperability Services§ Conclusions

Set your title here in Slide Master Options

Bringing Semantic Interoperability

7

{ "tmp" : "-2ºC","power" : "10Kw",”content": [

{ "type" : "apple"},{ "type" : "apple"},{ "type" : "orange"},

...] }

File storageJSON

Heterogeneus AccessHeterogeneus FormatsHeterogeneus Models

Heterogeneity:

Set your title here in Slide Master Options

Bringing Semantic Interoperability

8

{ "tmp" : "-2ºC","power" : "10Kw",”content": [

{ "type" : "apple"},{ "type" : "apple"},{ "type" : "orange"},

...] }

File storageJSON

Homogeneus AccessHomogeneus FormatHeterogeneus Models

Syntactic Interoperability:

Heterogeneus AccessHeterogeneus FormatsHeterogeneus Models

Heterogeneity:

REST APIRDF

Set your title here in Slide Master Options

Bringing Semantic Interoperability

9

{ "tmp" : "-2ºC","power" : "10Kw",”content": [

{ "type" : "apple"},{ "type" : "apple"},{ "type" : "orange"},

...] }

File storageJSON

Homogeneus AccessHomogeneus FormatHeterogeneus Models

Syntactic Interoperability:

Heterogeneus AccessHeterogeneus FormatsHeterogeneus Models

Heterogeneity:

REST APIRDF

Homogeneus AccessHomogeneus FormatHeterogeneus Models

Syntactic Interoperability: RDF

REST API sarefWoT

sosa…

Set your title here in Slide Master Options

HeterogeneusAccess,

Format, Model

Overcoming heterogeneity

10

Model1

Service

Datastore

<<exposes data>>

Set your title here in Slide Master Options

HeterogeneusAccess,

Format, Model

Overcoming heterogeneity

11

<<described by>>Model1

Service

Datastore

WoT

WoTMappings

<<exposes data>>

Set your title here in Slide Master Options

HeterogeneusAccess,

Format, Model

Overcoming heterogeneity

12

<<described by>>Model1

Service

Datastore

WoT

WoTMappings

Client

SPARQL Resources Dataset

<<reads>><<reads>> <<reads>>

HomogeneusAccess,

Format, Model

<<exposes data>>

Set your title here in Slide Master Options

Ontology overview

13

07-VICINITY Ontology Slide 7

Datatypes

WebofThings

WoTmappings

VICINITYCore

VICINITYAdapters

Set your title here in Slide Master Options

The WoT-Mappings

14

Set your title here in Slide Master Options

Semantic Interoperability Services: Registering “Things”

<<described by>>

<<described by>>

<<described by>>

<<described by>>

Description Repository

<<stored at>>

<<stored at>>

<<stored at>>

<<stored at>>

Description content:- Description- Contextual data- Mappings

How to translate non-RDF data into RDF

Set your title here in Slide Master Options

Semantic Interoperability Services: Discovery “Things”

DiscoveryServices

Agent

Client

Description Repository

<<discovery>>Onlysuitable

descriptionsareretrievedinthe

discovery

QueryResults

Ifquerydoesnotrequiretoaccessdata,theanswer

isprovided

1

2

3

4

Set your title here in Slide Master Options

Semantic Interoperability Services: Distributed access to “Things”

Agent

Client

QueryResults

Afterdiscovery,Thingsendpointsareretrievedand

accessed1

Endpoint 1 Endpoint 2 Endpoint 3…

<<access>>

Onlysuitableendpointsareaccessed2

3

Description andfetcheddataarecombinedtoanswerthe

query

Homogeneus AccessHomogeneus FormatHomogeneus Model

Set your title here in Slide Master Options

Semantic Interoperability Services: Distributed access to “Things”

Agent

Client

QueryResults

Afterdiscovery,Thingsendpointsareretrievedand

accessed1

Endpoint 1 Endpoint 2 Endpoint 3…

Heterogeneus AccessHeterogeneus FormatsHeterogeneus Models

<<access>>Onlysuitableendpointsareaccessed.Thus

theClientimplementsmechanismstoovercometheheterogeneousaccessof

endpoints(specifiedinmappings)

2

3

Translatesfetcheddatainto

RDFusingmappings

4

Set your title here in Slide Master Options

WoT in our research

19

WoT-Implementation

Projects:

Specifications:

Impementations:

Helio(mapping processor)

WoT-Mappings

Set your title here in Slide Master Options

Conclusions of our approach

20

§ Allows to transparently discover Things relaying on their descriptions

§ Performs a distributed access, if required, only to the suitable endpoints to answer a query

§ Provides a transparent mechanism to interact with heterogeneous environments of datao Mapping-based normalisation of access points,

format, model§ May follow any of the privacy policies required § Supports centralised or decentralised

architectures§ Based on WoT + WoT-mappings ontology

o Plus other ancillary ontologies like core, adapters, saref, …

Set your title here in Slide Master Options

Thanks to all

21

Achieving interoperability is not only about descriptions, butalso about providing mechanisms for automatic datadiscovery and access from heterogeneous data sources