There is no clear definition for the term knowledge graph. It is made well-known by Google with its initiative by the same name in 2012. Very often, people use the term to represent ontology in general. In some contexts, the term is used to refer to any knowledge base that is represented as a graph.
This post gives a pretty good introduction.
A demonstration of using publicly available data to create a simple knowledge graph. To be more precise, we do entity analysis based on articles published by our main interest (query) to form a network graph in order to organize information into knowledge (understanding).
The general idea is as follows,
- Find an information source
- Extract data
- Do entities analysis
- Plot network graph
Configuration (config.yml) is imported through package config. Sample file is provided.
We use data extracted from Zhihu.com (知乎), a platfrom of information sharing and article publishing in China. The following shows query of - Lighthouse Capital (光源资本)
Image via https://medium.com/@sderymail/challenges-of-knowledge-graph-part-1-d9ffe9e35214