How to use the sum service:
sum?entity=http://dbpedia.org/resource/Douglas_Adams&topK=10
sum?entity=http://dbpedia.org/resource/Original_Sin_(2001_film)&topK=10
sum?entity=http://dbpedia.org/resource/Usain_Bolt&topK=10
sum?entity=http://dbpedia.org/resource/Marie_Curie&topK=10
sum?entity=http://dbpedia.org/resource/Manchester_City_F.C.&topK=10
sum?entity=http://dbpedia.org/resource/Lexus&topK=10
sum?entity=http://dbpedia.org/resource/The_Cosby_Show&topK=10
curl -i -d "[ a <http://purl.org/voc/summa/Summary> ; \ <http://purl.org/voc/summa/entity> \ <http://dbpedia.org/resource/Porsche> ; \ <http://purl.org/voc/summa/topK> '5'; \ <http://purl.org/voc/summa/maxHops> '1'; \ <http://purl.org/voc/summa/language> 'en'; \ <http://purl.org/voc/summa/fixedProperty> \ <http://dbpedia.org/ontology/keyPerson>; \ <http://purl.org/voc/summa/fixedProperty> \ <http://dbpedia.org/ontology/owner> ] ." \ -H "content-type: text/turtle" -H "accept: application/ld+json" http://km.aifb.kit.edu/services/link/sum
Please cite LinkSUM in your publication as follows:
@InCollection{linksum, Title = {{LinkSUM: Using Link Analysis to Summarize Entity Data}}, Author = {Thalhammer, Andreas and Lasierra, Nelia and Rettinger, Achim}, Booktitle = {Web Engineering: 16th International Conference, ICWE 2016, Lugano, Switzerland, June 6-9, 2016. Proceedings}, Publisher = {Springer International Publishing}, Year = {2016}, Address = {Cham}, Pages = {244--261}, Series = {Lecture Notes in Computer Science}, Volume = {9671}, Doi = {10.1007/978-3-319-38791-8_14}, ISBN = {978-3-319-38791-8}, Url = {http://dx.doi.org/10.1007/978-3-319-38791-8_14} }
The configuration of this LinkSUM instance uses the following setup:
It is directly hooked to the official DBpedia SPARQL endpoint. The main SPARQL query is as follows:PREFIX vrank:<http://purl.org/voc/vrank#> PREFIX rdfs:<http://www.w3.org/2000/01/rdf-schema#> PREFIX dbp-ont:<http://dbpedia.org/ontology/> PREFIX dbp-prop:<http://dbpedia.org/property/> SELECT ?s ?p ?o ?slabel ?plabel ?olabel ?v FROM <http://dbpedia.org> FROM <http://people.aifb.kit.edu/ath/#DBpedia_PageRank> WHERE { { <http://dbpedia.org/resource/Douglas_Adams> ?p ?o. FILTER regex(str(?o),"http://dbpedia.org/resource","i"). FILTER (?p != dbp-ont:wikiPageWikiLink && ?p != <http://purl.org/dc/terms/subject> && ?p != dbp-prop:wikiPageUsesTemplate && ?p != rdfs:seeAlso && ?p != <http://www.w3.org/2002/07/owl#differentFrom>). OPTIONAL {?o rdfs:label ?olabel. FILTER langmatches( lang(?olabel), "EN" ). }. OPTIONAL {?p rdfs:label ?plabel. FILTER langmatches( lang(?plabel), "EN" ).}. OPTIONAL {<http://dbpedia.org/resource/Douglas_Adams> rdfs:label ?slabel. FILTER langmatches( lang(?slabel), "EN" ).}. OPTIONAL {?o vrank:hasRank ?r. ?r vrank:rankValue ?v}. } UNION { ?s ?p <http://dbpedia.org/resource/Douglas_Adams>. FILTER regex(str(?s),"http://dbpedia.org/resource","i"). FILTER (?p != dbp-ont:wikiPageWikiLink && ?p != <http://purl.org/dc/terms/subject> && ?p != dbp-prop:wikiPageUsesTemplate && ?p != rdfs:seeAlso && ?p != <http://www.w3.org/2002/07/owl#differentFrom>). OPTIONAL {?s rdfs:label ?slabel. FILTER langmatches( lang(?slabel), "EN" ). }. OPTIONAL {?p rdfs:label ?plabel. FILTER langmatches( lang(?plabel), "EN" ).}. OPTIONAL {<http://dbpedia.org/resource/Douglas_Adams> rdfs:label ?olabel. FILTER langmatches( lang(?olabel), "EN" ).}. OPTIONAL {?s vrank:hasRank ?r. ?r vrank:rankValue ?v}. } }For the backlinks, it also uses the Wikipedia API (as the intersection of the following two sets):