Question Answering over Knowledge Graphs (KGQA) aims to compute answers for
natural language questions over a knowledge graph. Recent KGQA approaches adopt
a neural machine translation (NMT) approach, where the natural language
question is translated into a structured query language. However, NMT suffers
from the out-ofvocabulary problem, where terms in a question may not have been
seen during training, impeding their translation. This issue is particularly
problematic for the millions of entities that large knowledge graphs describe.
We rather propose a KGQA approach that delegates the processing of entities to
entity linking (EL) systems. NMT is then used to create a query template with
placeholders that are filled by entities identified from the text in an EL
phase. This approach gives rise to what we call the "entity filling" problem,
where we must decide which placeholders to replace with which entities. To
address this problem, we propose a solution based on sequence labelling and
constraints. Experiments for QA with complex questions over Wikidata show that
our approach outperforms pure NMT approaches: while the task remains
challenging, errors relating to entities in the translated queries are greatly
reduced.