Entity Summarization in Knowledge Graphs: Algorithms, Evaluation, and Applications

Gong Cheng, Kalpa Gunaratna, Evgeny Kharlamov

Abstract: Knowledge graphs (KGs) encapsulate entities and relationships that describe the entities. The concise representation format and graph nature of KGs have resulted in creating many novel Web and industrial applications and enhancing existing ones. However, in a KG, dozens or hundreds of facts describing an entity could exceed the capacity of a typical user interface and overload users with excessive amounts of information. This has motivated fruitful research on entity summarization—automated generation of compact summaries for entities to satisfy users’ information needs efficiently and effectively. Over the recent years, researchers have contributed to this problem by proposing approaches ranging from pure ranking and mining techniques to machine and deep learning techniques. The state of the art has continuously improved and at the same time made it harder for the community and new comers to the problem to keep up with the recent contributions and basic building blocks in the space. This tutorial aims to fill this gap.