SPARQL应用实例

注1：以下所有代码块均可直接在https://dbpedia.org/sparql中运行。
注2：SPARQL对大小写不敏感。
前缀汇总：

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

查询满足条件的三元组数量。

select count(*) as ?total_num
where {
    ?s <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> ?name.
}

查询DBpedia中Class数量。

SELECT COUNT(DISTINCT ?class_name) AS ?total_num
WHERE {
    ?class_name rdf:type owl:Class
}

查询所有存在的Class。

SELECT *
WHERE {
    ?class_name rdf:type owl:Class
}

查询所有存在实体的Class类别。

SELECT (COUNT(DISTINCT ?Class_name) AS ?Class_num)
WHERE {
    ?Class_name rdf:type owl:Class.
    ?entity_name rdf:type ?Class_name.
}

查询所有存在实体的主类（rdf:type为owl:Thing）。

SELECT ?main_class (COUNT(DISTINCT ?entity_name) AS ?entity_num)
WHERE{
    ?main_class rdf:type owl:Class.
    ?main_class rdfs:subClassOf owl:Thing.
    ?entity_name rdf:type ?main_class.
}
GROUP BY ?main_class
ORDER BY desc (?entity_num)

查询每个Class及其实体数量并按照实体数量降序排序（此查询返回的Class类别不包括不含有实体的Class类别）。

SELECT ?Class_name (COUNT(?entity_name) AS ?entity_num)
WHERE {
    ?Class_name rdf:type owl:Class.
    ?entity_name rdf:type ?Class_name.
}
GROUP BY ?Class_name
ORDER BY desc (?entity_num)

查询Work类中的实体数量。

SELECT *
WHERE {
    ?entity_name rdf:type dbo:Work.
}

或（其余前缀均可按照类似方式使用。）

SELECT COUNT(DISTINCT ?entity_name)
WHERE {
    ?entity_name rdf:type <http://dbpedia.org/ontology/Work>.
}

查询Work类中具备英文摘要的实体数量。

SELECT COUNT(DISTINCT ?entity_name)
WHERE {
    ?entity_name rdf:type <http://dbpedia.org/ontology/Work>.
    ?entity_name dbo:abstract ?abstract.
    FILTER(LANGMATCHES(LANG(?abstract), "en"))
}

查询Work类中不具备英文摘要的实体数量（注意语言匹配中用"en"或"EN"均可）。

SELECT COUNT(DISTINCT ?entity_name)
WHERE {
    ?entity_name rdf:type <http://dbpedia.org/ontology/Work>.
    OPTIONAL{
        ?entity_name dbo:abstract ?abstract.
        FILTER(LANGMATCHES(LANG(?abstract), "en"))
    }
    FILTER(!bound(?abstract))
}

查询每个Class中具有英文摘要的实体数量（此查询所需时间较长，建议将SPARQL嵌入代码，每次查询仅对一个Class执行）。

SELECT ?Class_name (COUNT(DISTINCT ?entity_name) AS ?entity_num_with_en_abstract)
WHERE {
    ?Class_name rdf:type owl:Class.
    ?entity_name rdf:type ?Class_name.
    ?entity_name dbo:abstract ?abstract.
    FILTER(LANGMATCHES(LANG(?abstract), "en"))
}
GROUP BY ?Class_name
ORDER BY desc (?entity_num_with_en_abstract)

综合应用实例

from SPARQLWrapper import SPARQLWrapper, JSON
import json
sparql = SPARQLWrapper("http://dbpedia.org/sparql")
sparql.setQuery("""
    SELECT ?main_class (COUNT(DISTINCT ?entity_name) AS ?entity_num)
    WHERE{
        ?main_class rdf:type owl:Class.
        ?main_class rdfs:subClassOf owl:Thing.
        ?entity_name rdf:type ?main_class.
    }
    GROUP BY ?main_class
    ORDER BY desc (?entity_num)
""")
sparql.setReturnFormat(JSON)
sparql.setTimeout(9223372036)
results = sparql.query().convert()
class_dict = {}
for result in results["results"]["bindings"]:
    if 'http://dbpedia.org/ontology/' in result['main_class']['value']:
        class_dict[result['main_class']['value']] = -1
f = open('/home/njucs/Desktop/class_dict_keys.txt', 'w')
for item in class_dict.keys():
	print(item.ljust(50))
	print(item.ljust(50), file = f)
f.close()
f = open('/home/njucs/Desktop/class_dict.txt', 'w')
seq = 0
for key in class_dict.keys():
    sparql = SPARQLWrapper("http://dbpedia.org/sparql")
    sparql.setQuery("""
        SELECT (COUNT(DISTINCT ?entity_name) AS ?num)
        WHERE {
            ?entity_name rdf:type <""" + '{0}'.format(key) + """>.
            ?entity_name dbo:abstract ?abstract.
            FILTER(LANGMATCHES(LANG(?abstract), "en"))
        }
    """)
    sparql.setReturnFormat(JSON)
    sparql.setTimeout(9223372036)
    results = sparql.query().convert()
    class_dict[key] = int(results["results"]["bindings"][0]["num"]["value"])
    seq += 1
    print(seq, "\t", key.ljust(50), class_dict[key])
    print(key, class_dict[key], file=f)
f.close()
json_str = json.dumps(class_dict)
f = open('/home/njucs/Desktop/class_dict.json', 'w')
f.write(json_str)
f.close()
keys = list(class_dict.keys())
values = list(class_dict.values())
ordered_list = list(zip(keys, values))
ordered_list = sorted(ordered_list, key = (lambda x:x[1]), reverse = True)
json_str = json.dumps(ordered_list)
f = open('/home/njucs/Desktop/ordered_list.json', 'w')
f.write(json_str)
f.close()