Improving question answering over incomplete knowledge graphs with relation prediction

Large-scale knowledge graphs (KGs) play a critical role in question answering over KGs (KGs-QA). Despite of large scale, KGs suffer from incompleteness, which has fueled a lot of research on relation prediction. Since existing researches of relation prediction process each triple independently, the hidden relations which are inherently present can not be captured. Complementarily, to simultaneously capture both entity features and relation features in a given entity’s neighborhood, an entity importance estimation network of attention-based graph embedding is proposed, which consists of the attention-based graph embedding module and the entity importance estimation module. Firstly, the new embedding of an entity from its n-hop neighbor is learned by an attention-based graph embedding module. Then, the learned new embedding is integrated into the entity importance estimation module to find entities of high importance in n-hop neighbors of the central entity. Finally, multi-hop relations are encapsulated and an auxiliary edge of n-hop neighbors is introduced, which realizes the relation prediction task. To the best our knowledge, we are the first to realize KGs-QA while realizing relation prediction, which alleviates the phenomenon of missing relations and the low-precision problem of KGs-QA. On the SQ datasets, the proposed method obtains a high F1 score (49.3%) in 10% missing relation, compared to QASE and MCCNNs with F1 scores of 44.2% and 46.3%, respectively.