Rome Kres Hackathon

From STLab

Revision as of 16:34, 5 March 2011

Link to the work flow: http://stlab.istc.cnr.it/documents/events/kres-hackathon/rome-kres-hackaton-2.png

Link to the photogallery: KReSHackathonRomeGallery

LOD resources endpoint is : http://150.146.88.63:8893/sparql

Stanbol instance at : http://150.146.88.63:9292/

Detect dbpedia entities

Load enahcnements :: <stanbol>/store

1. Create a file with my example text:
   • movie.txt
   • "Una nuova missione per l'agente Ethan Hunt ( Tom Cruise ) che, questa volta, dovrà cercare di fermare la diffusione di Chimera, un virus geneticamente modificato che produce effetti letali sul sistema nervoso dell'essere umano, capace di mettere in ginocchio il mondo. Al fianco di Ethan il genio dell'informatica Luther Stickell ( Ving Rhames ) e l'affascinante ladra internazionale Nyah Nordoff-Hall (Thandie Newton). Il nemico da combattere, invece, è il perfido Sean Ambrose ( Dougray Scott ) il cui piano è di vendere su scala mondiale al miglior offerente l'antidoto al virus."
2. curl -i -X PUT -H "Content-Type:text/plain" -T movie1.txt http://150.146.88.63:9292/store/content/movie1.txt
3. Result: http://150.146.88.63:9292/store/page/movie1.txt

Select entity references :: <lod-resource>/sparql

prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> 
prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> 
prefix owl: <http://www.w3.org/2002/07/owl#> 
prefix xsd: <http://www.w3.org/2001/XMLSchema#> 

select distinct 
  ?z 
WHERE
{ 
    ?x <http://fise.iks-project.eu/ontology/extracted-from>  <http://150.146.88.63:9292/store/content/movie1.txt> .
    ?x <http://fise.iks-project.eu/ontology/entity-reference> ?z
}

OR

prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> 
prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> 
prefix owl: <http://www.w3.org/2002/07/owl#> 
prefix xsd: <http://www.w3.org/2001/XMLSchema#> 
construct {
 ?z rdf:type owl:Thing
}
WHERE
{ 
   ?x <http://fise.iks-project.eu/ontology/extracted-from>  <http://150.146.88.63:9292/store/content/movie1.txt> .
   ?x <http://fise.iks-project.eu/ontology/entity-reference> ?z
}

Select owl:sameAs from each LOD source

Linked MDB

PREFIX oddlinker: <http://data.linkedmdb.org/resource/oddlinker/>
CONSTRUCT {
 ?Subject ?type ?source
}
FROM <lmdb_dataset>

WHERE
{
?x oddlinker:link_source ?source .
?x oddlinker:link_type ?type .
?x oddlinker:link_target ?Subject .
FILTER ( ?type = "owl:sameAs" || ?type = "rdfs:SeeAlso") .
FILTER ( 
    ?Subject = <http://sws.geonames.org/2118968/> ||
    ?Subject = <http://dbpedia.org/resource/Tom_Cruise> ||
    ?Subject = <http://sws.geonames.org/556951/> ||
    ?Subject = <http://dbpedia.org/resource/Al_Pacino> ||
    ?Subject = <http://sws.geonames.org/106281/> ||
    ?Subject = <http://dbpedia.org/resource/Al_Gore>
 )
}
LIMIT 10000

And the result is:

Compute owl:sameAs with the reasoner :: /enrich

Fetch the information from LOD sources

SELECT *
FROM <lmdb_dataset>
WHERE{
 {<http://dbpedia.org/resource/Al_Gore> ?p ?o} 
 UNION 
 {?s ?pp <http://dbpedia.org/resource/Al_Gore>}
 UNION
 {<http://dbpedia.org/resource/Al_Pacino> ?p ?o} 
 UNION 
 {?s ?pp <http://dbpedia.org/resource/Al_Pacino>}
 UNION
 {<http://dbpedia.org/resource/Tom_Cruise> ?p ?o} 
 UNION 
 {?s ?pp <http://dbpedia.org/resource/Tom_Cruise>}
 UNION
 {<http://data.linkedmdb.org/resource/actor/12737> ?p ?o} 
 UNION 
 {?s ?pp <http://data.linkedmdb.org/resource/actor/12748>}
 UNION
 {<http://data.linkedmdb.org/resource/actor/12737> ?p ?o} 
 UNION 
 {?s ?pp <http://data.linkedmdb.org/resource/actor/12748>}
 UNION 
 {?s ?pp <http://data.linkedmdb.org/resource/actor/28241>}
 UNION
 {<http://data.linkedmdb.org/resource/actor/28241> ?p ?o} 
}

Select the Recipe and do /refactor

google = <http://rdf.data-vocabulary.org/#> . 
mdb = <http://data.linkedmdb.org/resource/movie/> . 
dcterms = <http://purl.org/dc/terms/> . 
foaf = <http://xmlns.com/foaf/0.1/> . 
 
!film2product[
	is(mdb:Film, ?film)
		->
	is(google:Product, ?film)
] . 
 
!actor2person[
	is(mdb:Actor, ?actor)
		->
	is(google:Person, ?actor) . 
	values(google:title, ?actor, "Actor")
] . 
 
actorname[
	is(google:Person, ?p) . 
	values(mdb:actor_name, ?p, ?name) . 
		->
	values(google:name, ?p, ?name)
] . 
 
actorCharacter[
	is(google:Person, ?p) . 
	has(mdb:performance, ?p, ?performance) .
	has(mdb:performance_character, ?performance, ?character) . 
	values(performance_film, ?performance, ?film) .
		->
	values(google:role, ?character) . 
	values(google:affiliation, ?film)
] . 
 
actorPage[
	is(google:Person, ?p) . 
	has(foaf:page, ?p, ?page) . 
	let(?pageString, str(?page))
		->
	values(google:role, ?pageString)
]