A knowledge graph organizes real world knowledge as entities and relationships between entities. Creating a knowledge graph often involves scraping / ingesting unstructured data and creating structure out of it by extracting entities and relationships automatically. Examples of knowledge graphs are Wordnet, DBpedia,..
A database also is a knowledge graph in some sense. Since it realizes edges between different nodes (through foreign keys). But a graph data structure is useful for many operations not realised efficiently through a typical db join. Example: Get all my friends in my 2-deep neighbourhood and aggregate their preferences of color. In a typical relational database, relationships are defined at a table level, while in a graph database, there is more flexibility and relationships can be unique to a node.