< Back to Glossary
Automated Knowledge Bases
are large repositories of knowledge structured as entities and the relationships between them that are compiled through some automated effort. While manually compiled and updated knowledge bases have long been prevalent within many contexts (marketing, support, IT, operations), the key differentiator is that automated knowledge bases typically rely on machine learning to constitute themselves, ensure accuracy, and grow.
A key difference between a knowledge graph and a knowledge base comes down the notion of a graph. All knowledge graphs are knowledge bases, but not all knowledge bases are created in the form of graphs. Three key differentiators between knowledge graphs and knowledge bases include:
- Knowledge graphs typically provide a greater focus on connections between entities
- Knowledge graphs are semantic representations. This means data is encoded side-by-side with the meaning of the data.
- Knowledge graphs rely on an ontology
There are certainly knowledge bases which also focus on connections and semantic data (particularly automated knowledge bases). With that said, a knowledge base is typically thought of as a wider term that also constitutes instances such as help desk resources, a collection or articles, or arrangements of media meant for purposes other than a database.