Re-ranking Entities. Training the Wikipedia2Vec model on a Wikipedia knowledge graph results in a single graph embedding vector for every Wikipedia entity. The next question to answer is how to use these graph embeddings in the setting of entity retrieval. We propose a two-stage ranking model, where we first produce a ranking of candidate entities using state-of-the-art entity retrieval models (see Sect. 2.2), and then use the graph embeddings to reorder these entities based on their similarity to the query entities, as measured in the derived graph embedding space. Following the related work discussed in Sect. 2.3, we use the Tagme entity linker to identify the entities mentioned in the query. Given input query Q, we obtain a set of linked entities E(Q) and a confidence score s(e) for each entity, which represents the strength of the relationship between the query and the linked entity. We then compute an embedding-based score for every query Q Σ and entity E: F (E, Q)= e∈E(Q) s(e) · cos(−→E, −→e ), (5) where →−E, −→e denote the embeddings vectors for entities E and e. The rationale for this approach is the hypothesis that relevant entities for a given query are situated close (in graph embedding space) to the query entities identified by the entity linker. Consider for example the query “Who is the daughter of Xxxx Xxxxxxx married to.” Tagme links the query to entities Xxxx Xxxxxxx with a confidence of 0.66, Daughter with a confidence of 0.13, and Same-sex marriage with a confi- dence score of 0.21. Highly ranked entities then have a large similarity to these entities, where similarity to Xxxx Xxxxxxx adds more to the score than similarity to Daughter or Same-sex marriage (as the confidence score of Xxxx Xxxxxxx is higher than the other two). The relevant entities for this query (according to the DBpedia-Entity V2 test collection [12]) are Xxxxxxx Xxxxxxx, who is Xxxx Xxxx- ton’s daughter, and Xxxxxxx Family. We can reasonably expect these entities to have similarity to the linked entities, confirming our intuition.
