Knowledge Graph
Knowledge Graph
A data structure that connects various pieces of knowledge with nodes and edges, making their relationships visible.
In Simple Terms
A knowledge graph is a system that shows how scattered pieces of information connect to one another. For example, linking 'Natsume Soseki' as a node and 'Kokoro' as a node with an edge labeled 'author' makes their relationship immediately clear. This technology is used in features like Google Search, where looking up a person's name displays a summary of their biography and related works alongside the results.
Behind the Name
The name 'Knowledge Graph' combines two words: Knowledge and Graph. Here, 'graph' doesn't refer to a bar chart or pie chart — it comes from a branch of mathematics called 'graph theory,' which uses nodes (points) and edges (lines) to represent connections between things. The name reflects the idea of organizing vast amounts of real-world knowledge using this graph structure.
Take a Closer Look!
A knowledge graph is a data structure that links various pieces of information and knowledge in the world as 'things' and 'relationships,' forming a network.
By accumulating a large number of connections — such as 'A is B' or 'A is the author of B' — computers can organize and process complex knowledge more effectively.
While traditional databases are often organized in table form, a knowledge graph is characterized by its web-like, interconnected structure.
This makes it easier for AI to discover unexpected links between two pieces of information that are not directly connected.
Knowledge graphs are widely used in search engines, product recommendation systems, and voice assistant interactions.
They serve as a crucial foundation that enables AI to not just scan text, but to 'understand' the meaning and context behind words when generating responses.