CounQER

Method

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

1. identify set-valued predicates from a given KB predicates via statistical and embedding-based features, and
2. 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,

1. Wikidata
2. DBpedia raw extractions
3. DBpedia mapped extractions
4. Freebase

We 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.

CounQER Output