Icons made by Freepik from www.flaticon.com is licensed by CC 3.0 BY

Knowledge Graph

A knowledge graph can be intuitively described as an encyclopedia for machines. It contains knowledge formatted in a way that can be understood and reasoned about by computers. The Ambiverse knowledge graph consists of three main elements: named entities, categories and facts.

Named Entities

Named entities, the backbone of our knowledge graph, are (physical or abstract) real world objects like persons, organizations, locations, fictional characters, products, etc. Bob Dylan, PixarUnited StatesBuzz LightyearGibson Guitar or the Nobel Prize are examples of named entities.

 
By Alberto Cabello from Vitoria Gasteiz - Bob Dylan, CC BY 2.0, Link
By Source, Fair use, Link
By Elkman - Own work, CC BY-SA 3.0, Link
By Source, Fair use, Link

Categories

A useful aspect about named entities is that they are categorizable. They can be grouped into persons, locations, organizations, products, or any other category and each category, in turn, can be further subcategorized. For example, a person can be a singer like Bob Dylan or an inventor like Steve Jobs; a product can be an album (e.g., Highway 61 Revisited) or a guitar (e.g., Gibson Hummingbird). This forms the hierarchical tree in which the nodes are categories, and the leaves are named entities.

Facts

A fact is something that is known to be true; It is expressed by linking named entities with relations. For instance, <Bob Dylan> <wasBornIn> <Duluth> is a fact consisting of two named entities, Bob Dylan and Duluth, and the relation wasBornIn. Facts can also link quantities or dates to entities, like wasBornOnDate, which states the birth date of a certain person.

Multi-Lingual Knowledge

Our knowledge graph contains entity names in English, but also Chinese, German, and Spanish.

Linked Data

The Ambiverse knowledge graph is connected to the linked data community allowing interconnections to a large number of other knowledge-graphs.

 
By Max Schmachtenberg, Christian Bizer, Anja Jentzsch and Richard Cyganiak - http://lod-cloud.net/, CC BY-SA 3.0, Link

The Ambiverse Knowledge Graph

The Ambiverse knowledge graph is based on YAGO, a high precision knowledge base developed for over a decade at the Max Planck Institute for Informatics. YAGO automatically transforms Wikipedia into a computer readable form. Ambiverse provides this general purpose out-of-the- box knowledge graph which can be integrated and enriched with customer-specific knowledge.   YAGO in numbers 

Named Entities 10.5M
Categories 650K
Facts 16M
Learn more about YAGO here. YAGO

Custom Knowledge Graphs

Ambiverse technology allows the generation of customer-specific knowledge graphs from customer internal data sources (e.g., databases, text documents, structured sources, unstructured sources, etc). This knowledge can be further integrated with our or third party knowledge graphs relevant for the customer (e.g., financial information, customer information, product information, etc).