Overview of Knowledge Graph Completion Algorithms Based on Representation Learning

In recent years, the KG (Knowledge graph), as a new generation of knowledge engineering technical representatives, has been developed vigorously. KG is a key step for artificial intelligence to move towards intelligence. Knowledge graph completion (KGC) is a key technology in the construction of KG. The aim is to complement the missing knowledge in the KG. With the development of the Internet, the scale of data increases exponentially, and the scale of KG is increasing day by day. It is very inefficient to make up the missing knowledge based on manual, so many algorithms of KGC have emerged. In this paper, some typical KGC algorithms based on representation learning are summarized and combed, which are mainly divided into tensor decomposition, neural network and translation transformation. The development direction and prospect of KGC in KG are discussed and prospected.