Explore Entity Embedding Effectiveness in Entity Retrieval

Abstract

This paper explores entity embedding effectiveness in ad-hoc entity retrieval, which introduces distributed representation of entities into entity retrieval. The knowledge graph contains lots of knowledge and models entity semantic relations with the well-formed structural representation. Entity embedding learns lots of semantic information from the knowledge graph and represents entities with a low-dimensional representation, which provides an opportunity to establish interactions between query related entities and candidate entities for entity retrieval. Our experiments demonstrate the effectiveness of entity embedding based model, which achieves more than 5% improvement than the previous state-of-the-art learning to rank based entity retrieval model. Our further analysis reveals that the entity semantic match feature effective, especially for the scenario which needs more semantic understanding.

Joint Word and Entity Embeddings for Entity Retrieval from a Knowledge Graph

“…While earlier studies [5,32,35] heavily utilized knowledge graph's structure during retrieval, more recent studies [21,22,25,45] only use it to construct fielded entity representations, effectively casting entity search into an instance of structured document retrieval. Entity similarity information obtained from entity embeddings was successfully utilized for re-ranking the results of termbased retrieval models in [14,17,44] using a learning-to-rank approach. A publicly available benchmark for entity search based on DBpedia [16] and its more recent version [13], which provides graded relevance judgments obtained using crowdsourcing and subsequent conflict resolution by experts, are standard test collections for evaluating entity search methods.…”

Graph-Embedding Empowered Entity Retrieval

“…Very recent work has applied Trans-E graph embeddings to the problem of entity retrieval, and shown consistent but small improvements [15]. However, Trans-E graph embeddings are not a good choice if the graph has 1-to-many, transitive or symmetric relations, which is the case in knowledge graphs [1].…”