Interopérabilité
Sur le schéma suivant, on peut voir une vue d'enssemble de tout le chantier sur l'intéropérabilité de Communecter avec divers sources.
A gauche on à la liste des sources extérieurs sur lesquelles on récupère les données :
- Wikidata
- Wikipédia
- OpenStreetMap
- OpenDataSoft (la base SIRENE)
- Data.gouv
- Datanova (les enseignes La Poste)
- Pôle Emploi
- SCANR
Au milieu, le processus de Convertion des données (détails sur la prochaine figure)
A droite, l'affichage des données converties sur le site de Communecter ainsi que des exemple d'usage deces données par des sites exterieurs.
Le schéma suivant décrit en détail la convertion des données sémantiques.
For any city, We retreive main information available on Wikidata
The process is the following :
- We choose a geographic scope (a country) to filter
- We call our own semantic convert system (doc avaible here) :
The convert system will interrogate the Wikidata API to get data in JSON.
The next exemple is the data for the city of Saint-Denis, capital city of Réunion island :
https://query.wikidata.org/sparql?format=json&query=SELECT%20DISTINCT%20%3Fitem%20%3FitemLabel%20%3FitemDescription%20%3Fcoor%20%3Frange%20WHERE%20{%0A%20%3Fitem%20wdt%3AP131%20wd%3AQ47045.%0A%20%3Fitem%20%3Frange%20wd%3AQ47045.%0A%20%3Fitem%20wdt%3AP625%20%3Fcoor.%0A%20SERVICE%20wikibase%3Alabel%20{%20bd%3AserviceParam%20wikibase%3Alanguage%20%22fr%22.%20}%0A}
And convert this data in the pivot language named "PH onthology"
/ph/api/convert/wikipedia?url=https://www.wikidata.org/wiki/Special:EntityData/Q47045.json
Here are the mapping
| Source's data | PH onthology |
|---|---|
| itemLabel.value | name |
| coor.latitude | geo.latitude |
| coor.longitude | geo.longitude |
| item.value | url |
| itemDescription.value | description |
- We'll want to contributing back any extra data we can offer with COpédia (coming soon)
- For any city, We retreive main information available on Wikipedia
For any city, we retreive main information avaible on OSM
The process is the following :
- We choose a geographic scope (a country) to filter
- We call our own semantic convert system (doc avaible here) :
The next exemple is all the OSM data of the city of Saint-Louis :
http://overpass-api.de/api/interpreter?data=[out:json];node[%22name%22](poly:%22-21.303505996763%2055.403919253998%20-21.292626813288%2055.391189163162%20-21.282029142394%2055.381522536523%20-21.256155186265%2055.392395046639%20-21.232012804782%2055.387888015185%20-21.211100938923%2055.390619722192%20-21.199480966855%2055.382654775478%20-21.185882138486%2055.385961778627%20-21.173346518752%2055.389949958731%20-21.16327583783%2055.399563417107%20-21.14709868917%2055.405379688232%20-21.166028899095%2055.414700890276%20-21.184085220909%2055.432085218794%20-21.190290936422%2055.440880800108%20-21.195166490948%2055.462318490892%20-21.237553168259%2055.459769285867%20-21.258726107298%2055.463692709631%20-21.286021128961%2055.455515913879%20-21.294777773557%2055.419916682666%20-21.303505996763%2055.403919253998%22);out%2030;
Here are the mapping
| Source's data | PH onthology |
|---|---|
| tags.name | name |
| lat | geo.latitude |
| long | geo.longitude |
| type | type |
| tags.amenity | tags.0 |
- We'll want to contributin back any extra data we can offer with COSM (coming soon)
For any city, we retreive main information of the organizations placed in this city
The process is the following :
- We choose a geographic scope (a country) to filter
- We call our own semantic convert system (doc avaible here) :
The module will find all the organizations placed in the geographic scope filter and then extract all the data in the differents datasets available.
The next exemple is all the data of the different structure of Méto-France, meteorological center of France.
https://www.data.gouv.fr/api/1/datasets/54a12162c751df720a04805a/
Here are the mapping
| Source's data | PH onthology |
|---|---|
| slug | name |
| page | url |
| tags[] | tag[] |
| item.value | url |
| owner | creator |
For any city, we retreive all the job offer. (no exact localisation of the job place)
The process is the following :
- We choose a geographic scope (a country) to filter
- We call our own semantic convert system (doc avaible here) :
To get data with the Pôle emploi's API, a token is needed.
The next exemple fetch all the job offer of the city of Saint-Louis.
https://api.emploi-store.fr/partenaire/infotravail/v1/datastore_search_sql?sql=SELECT%20%2A%20FROM%20%22421692f5-f342-4223-9c51-72a27dcaf51e%22%20WHERE%20%22CITY_CODE%22=%2797414%27%20LIMIT%2030
For any city, we retreive all the organizations and the association of the SIREN's database.
The process is the following :
- We choose a geographic scope (a country) to filter
- We call our own semantic convert system (doc avaible here) :
The next exemple will fetch all the data in the SIRENE database for the city of Saint-Louis.
https://data.opendatasoft.com/api/records/1.0/search/?dataset=sirene%40public&facet=categorie&facet=proden&facet=libapen&facet=siege&facet=libreg_new&facet=saisonat&facet=libtefen&facet=depet&facet=libnj&facet=libtca&facet=liborigine&rows=30&start=0&geofilter.polygon=(-21.303505996763,55.403919253998),(-21.292626813288,55.391189163162),(-21.282029142394,55.381522536523),(-21.256155186265,55.392395046639),(-21.232012804782,55.387888015185),(-21.211100938923,55.390619722192),(-21.199480966855,55.382654775478),(-21.185882138486,55.385961778627),(-21.173346518752,55.389949958731),(-21.16327583783,55.399563417107),(-21.14709868917,55.405379688232),(-21.166028899095,55.414700890276),(-21.184085220909,55.432085218794),(-21.190290936422,55.440880800108),(-21.195166490948,55.462318490892),(-21.237553168259,55.459769285867),(-21.258726107298,55.463692709631),(-21.286021128961,55.455515913879),(-21.294777773557,55.419916682666),(-21.303505996763,55.403919253998)
Here are the mapping
| Source's data | PH onthology |
|---|---|
| fields.l1_declaree | name |
| fields.categorie | type |
| fields.siret | shortDescription |
| fields.coordonnees.0 | geo.latitude |
| fields.coordonnees.1 | geo.longitude |
| fields.libapen | tags.0 |
For any city, we retreive main information from the national education of France
The process is the following :
- We choose a geographic scope (a country) to filter
- We call our own semantic convert system (doc avaible here) :
The next exemple fetch all the actives research strutures of the city of Bordeaux :
https://data.enseignementsup-recherche.gouv.fr/api/records/1.0/search/?dataset=fr-esr-etablissements-publics-prives-impliques-recherche-developpement&facet=siren&facet=libelle&facet=date_de_creation&facet=categorie&facet=libelle_ape&facet=tranche_etp&facet=categorie_juridique&facet=wikidata&facet=commune&facet=unite_urbaine&facet=departement&facet=region&facet=pays&facet=badge&facet=region_avant_2016&rows=30&start=0&geofilter.polygon=(44.810795852605,-0.5738778170842),(44.817148298105,-0.57643460444186),(44.823910193873,-0.58695822406613),(44.818476638462,-0.60304723869607),(44.822474304509,-0.61064859861704),(44.824937843733,-0.61415033833008),(44.835177466959,-0.61079419661495),(44.841384923705,-0.62771243191386),(44.860667021743,-0.63833642556746),(44.871658097695,-0.63105127891779),(44.86227970331,-0.61630176568479),(44.854215265872,-0.59460939385687),(44.865671076253,-0.57646019656194),(44.869188961886,-0.57608874140575),(44.909402227434,-0.58088555560083),(44.908480410411,-0.57648917779388),(44.916666965125,-0.54773554113942),(44.889099273803,-0.53553255107571),(44.869138522062,-0.54141014437767),(44.868086689933,-0.53680669655034),(44.861267174723,-0.53784686147751),(44.848134506953,-0.53761462401784),(44.842390488778,-0.5422310311368),(44.836291776079,-0.54665943781219),(44.829021270567,-0.53642317794196),(44.822772234064,-0.53766321563778),(44.813135278103,-0.55606047183132),(44.810795852605,-0.5738778170842)
Here are the mapping :
| Source's data | PH onthology |
|---|---|
| fields.libelle | name |
| fields.site_web | shortDescription |
| fields.geolocalisation.0 | geo.latitude |
| fields.geolocalisation.1 | geo.longitude |
- Datasets used :
- Public or private research and development structures
- Member of the university institute of France
For any city, we retreive the location of all buildings of La Poste
The process is the following :
- We choose a geographic scope (a country) to filter
- We call our own semantic convert system (doc avaible here) :
The next exemple will fetch all La Poste buildings localised in the city of Saint-Louis.
https://datanova.laposte.fr/api/records/1.0/search/?dataset=laposte_poincont&rows=30&start=0&geofilter.polygon=(-21.303505996763,55.403919253998),(-21.292626813288,55.391189163162),(-21.282029142394,55.381522536523),(-21.256155186265,55.392395046639),(-21.232012804782,55.387888015185),(-21.211100938923,55.390619722192),(-21.199480966855,55.382654775478),(-21.185882138486,55.385961778627),(-21.173346518752,55.389949958731),(-21.16327583783,55.399563417107),(-21.14709868917,55.405379688232),(-21.166028899095,55.414700890276),(-21.184085220909,55.432085218794),(-21.190290936422,55.440880800108),(-21.195166490948,55.462318490892),(-21.237553168259,55.459769285867),(-21.258726107298,55.463692709631),(-21.286021128961,55.455515913879),(-21.294777773557,55.419916682666),(-21.303505996763,55.403919253998)
Here are the mapping
| Source's data | PH onthology |
|---|---|
| fields.libelle_du_site | name |
| recordid | type |
| fields.adresse | address.streetAddress |
| fields.latlong.0 | geo.latitude |
| fields.latlong.1 | geo.longitude |
| fields.libapen | tags.0 |
- onclick : we'll show all SCK kits for a given city
- POI's of type geoJson, on click we show the content on our map
- any WP blog's RSS can be pluggued to an elements wall
- users can use Framapads from inside CO(simple Iframe)