Here we present the alignments found by CounQER in each of the four KBs.
Click on a tab to select a KB.
Under each tab is a table with alignments on each row and the set predicate type and alignment score in the columns.
The tables are searchable and sortable by the columns.
If a set predicate exists in the SPO dropdown but not in table of the corresponding KB, the said predicate has no alignment.
Try out the link to the SPARQL query interface made available for each alignment.
Each link opens with a ready-made query to look up entities which have the alignment.
We denote alignments missed by CounQER with a * by the scores.
Predicate constraints of general-purpose knowledge bases (KBs) like Wikidata, DBpedia and Freebase are often limited to subproperty, domain and range constraints. In this demo we showcase CounQER, a system that illustrates the alignment of counting predicates, like staffSize, and enumerating predicates, like workInstitution-1. Here, users can inspect these alignments through a simple SPO question answering interface.
CounQER is built as a two step approach where we
identify set-valued predicates from a given KB predicates via statistical and embedding-based features, and
link counting predicates and enumerating predicates by a combination of co-occurrence, correlation and textual relatedness metrics.
We analyze the prevalence of count information in four prominent knowledge bases,
DBpedia raw extractions
DBpedia mapped extractions
and show that our linking method achieves up to 0.55 F1 score in set predicate identification versus 0.40 F1 score of a random selection, and normalized discounted gains of up to 0.84 at position 1 and 0.75 at position 3 in relevant predicate alignments. You can refer to our paper for complete details on the CounQER methodology.