In Python, I am querying SPARQL queries using SPARQLWRAPPER. I currently have two queries: # A lot of PREFIX definitions at the top of the query INSERT {{ GRAPH <https://graph-dev.golemlab.eu/graph/Dennis-01> {{ gd:{recent_id} a lrm:F1 ; lrm:R3_is_realised_in gd:{recent_id}.{expression_id} ; schema:title ?titleWithLang ; crm130_shows_features_of ?fandomId . gd:{recent_id}.{expression_id} a lrm:F2_Expression ; skosrefLabel ?titleWithLang ; schema:language ?language . }} }} WHERE {{ GRAPH <http://example.com> {{ ?s golem:story_id {story_id} ; golem:title ?title ; golem:language ?language ; golem:fandom ?fandom . VALUES (?language ?langCode) {{ {lang_map_sparql} }} BIND(STRLANG(STR(?title), STR(?langCode)) AS ?titleWithLang) }} GRAPH <https://graph-dev.golemlab.eu/graph/Dennis-01> {{ ?fandomId a gc:G15_Fandom . OPTIONAL {{ ?fandomId skosrefLabel ?prefLabel . }} OPTIONAL {{ ?fandomId skos:alternativeLabel ?altLabel . }} }} BIND(COALESCE(?prefLabel, ?altLabel) AS ?fandomLabel) FILTER(?fandomLabel = ?fandom) }} and # A lot of PREFIX definitions at the top of the query INSERT {{ GRAPH <https://graph-dev.golemlab.eu/graph/Dennis-01> {{ gd:{recent_id}.{expression_id} schema:datePublished ?datePublishedNew ; schema:dateModified ?dateModifiedNew ; schema:endTime ?datePackagedNew . }} }} WHERE {{ GRAPH <http://example.com> {{ ?s golem:story_id {story_id} ; golem:datePublished ?datePublished ; golem:dateModified ?dateModified ; golem:datePackaged ?datePackaged . }} BIND(STRDT(REPLACE(STR(?datePublished)," ",""), xsd:date) AS ?datePublishedNew) . BIND(STRDT(REPLACE(STR(?dateModified)," ",""), xsd:date) AS ?dateModifiedNew) . BIND(STRDT(REPLACE(STR(?datePackaged)," ",""), xsd:dateTime) AS ?datePackagedNew) . }} When I query for the subgraph using: # A lot of PREFIX definitions at the top of the query WITH GRAPH <https://graph-dev.golemlab.eu/graph/Dennis-01> SELECT * WHERE {{ ?s ?p ?o . FILTER NOT EXISTS {{ ?s a gc:G15_Fandom }} }} I am filtering out ?s where the object is a Fandom, since there are thousands of fandom instances in my subgraphs. I only get the triples that related to the date (my second query). I am expecting more instances than just the second query. For example, I'm expecting triples like: W01 schema:title "title" Is there something that I am doing wrong? How could I fix this? Continue reading...