Workaround for *P31/*P279

[WolfgangFahl]

QLever is great and quick and helpful. Unfortunately there are still limit for quite general problems
E.g https://qlever.cs.uni-freiburg.de/wikidata/8FVzZI
doesn't work but a workaround would be helpful
What can be done with the currents state of Qlever?

[hannahbast]

That is not really a limit. If you specify a predicate path like wdt:P279*/wdt:P31*, with a * after both predicates, then according to the standard, the subject (?academic_conferenceItem in this case) should also match the object (wd:Q2020153 in this case), which is almost never what you want (wd:Q2020153 itself is not an academic conference).

The right pattern for such extended type queries is wdt:P31/wdt:P279*, and QLever's autocompletion even suggests this when you construct a query incrementally. The corresponding variant of your query is https://qlever.cs.uni-freiburg.de/wikidata/0O82PS and it works without problems.

The query does not make too much sense, however. I guess what you want is, the number of conferences per location. You get that with this query: https://qlever.cs.uni-freiburg.de/wikidata/RRBkgH . If you want the total number of conferences in the respective country in the same table (repeated for every location in that country), you need a more complicated query (you need to have two separate subqueries, one for the location count and one for the country count, and join them).