java - Sparql 查询是否重复我不明白

Question

我正在使用此查询来获取所有编程语言及其详细信息。这是我的测试课。我在 java 中使用过它并且工作正常。我面临的问题是，有一种语言名为“ML(编程语言)”

它以不同的抽象、不同的影响打印了多次。不仅是机器学习，其他一些语言也这样做。我不知道我的查询是否有任何问题，或者它按原样获取准确数据。

package io.naztech.dbpedia;

import java.io.ByteArrayOutputStream;
import java.util.List;

import org.apache.jena.query.ResultSet;
import org.apache.jena.query.ResultSetFormatter;
import org.apache.jena.sparql.engine.http.QueryEngineHTTP;
import org.junit.BeforeClass;
import org.junit.Test;

import io.naztech.talent.model.PediaTag;

public class testDataFetching {

    @Test
    public void testAllDataFetching() {

        String q =  "PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> \n"+
                    "PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#> \n"+
                    "PREFIX dbo: <http://dbpedia.org/ontology/> \n"+
                    "PREFIX dbp: <http://dbpedia.org/property/> \n"+
                    "PREFIX owl: <http://www.w3.org/2002/07/owl#> \n"+
                    "PREFIX xsd: <http://www.w3.org/2001/XMLSchema#> \n" +
                    "PREFIX foaf: <http://xmlns.com/foaf/0.1/> \n" +
                    "PREFIX dc: <http://purl.org/dc/elements/1.1/> \n" +
                    "PREFIX : <http://dbpedia.org/resource/> \n" +
                    "PREFIX dbpedia2: <http://dbpedia.org/property/> \n" +
                    "PREFIX dbpedia: <http://dbpedia.org/> \n" +
                    "PREFIX skos: <http://www.w3.org/2004/02/skos/core#> \n" +

                    "SELECT DISTINCT ?pl ?pl_label ?abstract ?_thumbnail \n" +
                    "( Group_concat ( DISTINCT ?_influenced_label; separator= \", \")   AS ?influenced ) \n" + 
                    "( Group_concat ( DISTINCT ?_influencedBy_label; separator= \", \") AS ?influencedBy ) \n" + 
                    "( group_concat ( DISTINCT ?_sameAs; separator=\", \" ) AS ?sameAs ) \n" +
                    "( group_concat ( DISTINCT ?_paradigm_label; separator=\", \" ) AS ?paradigm ) \n" +

                    "WHERE  {\n" +

                    "       ?pl rdf:type dbo:ProgrammingLanguage .\n" + 

                    "       OPTIONAL { ?pl dbo:abstract ?abstract .\n" + 

                    "       FILTER ( LANG ( ?abstract ) = 'en' ) . } \n" + 

                    "       ?pl rdfs:label ?pl_label .\n" + 

                    "       FILTER ( LANG ( ?pl_label ) = 'en' ) .\n" + 

                    "       OPTIONAL { ?pl dbo:influenced ?_influenced . \n" + 

                    "       ?_influenced rdfs:label ?_influenced_label . \n" + 

                    "       FILTER ( LANG ( ?_influenced_label ) = 'en' ) . } \n" + 

                    "       OPTIONAL { ?pl dbo:influencedBy  ?_influencedBy . \n" + 

                    "       ?_influencedBy  rdfs:label ?_influencedBy_label . \n" + 

                    "       FILTER ( LANG ( ?_influencedBy_label ) = 'en' ) . } \n" +

                    "       OPTIONAL { ?pl owl:sameAs ?_sameAs . } \n" +

                    "       OPTIONAL { ?pl dbp:paradigm ?_paradigm . \n" +

                    "       ?_paradigm rdfs:label ?_paradigm_label . } \n" + 

                    "       OPTIONAL { ?pl dbo:thumbnail ?_thumbnail . } \n" +

                    "       }"+

                    "       GROUP BY ?pl ?pl_label ?abstract ?_thumbnail ?influenced ?influencedBy ?sameAs ?paradigm";

        @SuppressWarnings("resource")
        QueryEngineHTTP queryEngine = new QueryEngineHTTP("http://live.dbpedia.org/sparql", q);
        ResultSet results = queryEngine.execSelect();

        int count = 0;

        while (results.hasNext()) 
        {
            QuerySolution qs =  results.next();
            System.out.println("NAME-->\n"+qs.get("pl_label").toString()+"\n");

            if(qs.get("influenced") != null)
            {
            System.out.println("INFLUENCED-->\n"+qs.get("influenced").toString()+"\n"); 
            }
           if(qs.get("influencedBy") != null)
            {
                System.out.println("INFLUENCED BY-->\n"+qs.get("influencedBy").toString()+"\n"); 
            }
           if(qs.get("abstract") != null)
            {
                System.out.println("ABSTRACT-->\n"+qs.get("abstract").toString()+"\n");  
            }

            if(qs.get("sameAs") != null)
            {
                System.out.println("SAME AS-->\n"+qs.get("sameAs").toString()+"\n");  
            }

            if(qs.get("paradigm") != null)
            {
            System.out.println("PARADIGM-->\n"+qs.get("paradigm").toString()+"\n");  
            }

            if(qs.get("_thumbnail") != null)
            {
                System.out.println("THUMBNAIL-->\n"+qs.get("_thumbnail").toString()+"\n");  
            }

            System.out.println("\n");

            count++;
        }

        System.out.println(count);



    }

}

Answer 1

数据集中有3个英文摘要，看DBpedia Live resource .

您可以通过从 group by ... 部分中删除 ?abstract 变量并使用聚合函数(sample, min, max ) 获取任何摘要:

SELECT  ?pl ?pl_label 
        (MIN(?_abstract) AS ?abstract) # <- used MIN here to ensure stable result
        ?_thumbnail 
        (GROUP_CONCAT(DISTINCT ?_influenced_label ; separator='; ') AS ?influenced) 
        (GROUP_CONCAT(DISTINCT ?_influencedBy_label ; separator='; ') AS ?influencedBy) 
        (GROUP_CONCAT(DISTINCT ?_sameAs ; separator=', ') AS ?sameAs) 
        (GROUP_CONCAT(DISTINCT ?_paradigm_label ; separator=', ') AS ?paradigm)
WHERE
  { ?pl  a  dbo:ProgrammingLanguage ;
         rdfs:label  ?pl_label
    FILTER ( lang(?pl_label) = "en" )

    OPTIONAL
      { ?pl  dbo:abstract  ?_abstract
        FILTER ( lang(?_abstract) = "en" )
      }
    OPTIONAL
      { ?pl       dbo:influenced/rdfs:label  ?_influenced_label
        FILTER ( lang(?_influenced_label) = "en" )
      }
    OPTIONAL
      { ?pl       dbo:influencedBy/rdfs:label  ?_influencedBy_label
        FILTER ( lang(?_influencedBy_label) = "en" )
      }
    OPTIONAL
      { ?pl  owl:sameAs  ?_sameAs }
    OPTIONAL
      { ?pl       dbp:paradigm/rdfs:label  ?_paradigm_label
        FILTER ( lang(?_paradigm_label) = "en" )
      }
    OPTIONAL
      { ?pl  dbo:thumbnail  ?_thumbnail }
  }
GROUP BY ?pl ?pl_label ?_thumbnail

更新

我将在此处添加@TallTed 的评论，他是 Virtuoso 背后的人之一，并且比我更了解它:

Be aware that while the recommended aggregate functions (MIN, MAX, SAMPLE) will get a value of the predicate, there is no assurance that this value will be the latest ingested to the dataset.