Knowledge Fusion: Introduction of Concepts and Techniques

The Internet contains a huge amount of data information, relying only on human resources can no longer obtain the complete and useful information from it. The research of knowledge graph provides a means to obtain large amounts of knowledge from massive texts and images, which plays an important role in people's daily life, such as QA, search, recommendation. The effective organization and integration of knowledge obtained from different sources into a knowledge base is the researching difficulty and hot spot to be solved in the knowledge graph construction in currently. This paper summarizes the various methods and research progress involved in knowledge fusion from three aspects, that is, open network knowledge fusion and multiple knowledge base knowledge fusion, and knowledge evaluation. We hope it can provide some useful guidance and inspiration for the research of knowledge fusion and knowledge graph construction of future network.