From the course: Hands-On AI: Knowledge Graphs for Generative AI Use Cases

Unlock the full course today

Join today to access over 24,900 courses taught by industry experts.

Sample architecture

Sample architecture

From the course: Hands-On AI: Knowledge Graphs for Generative AI Use Cases

Start my 1-month free trial Buy for my team

Sample architecture

“

- [Narrator] So let's tie all of this together now. The knowledge graph architecture can use a lot of components you likely have in your existing architecture, like your databases, file shares, ETL pipeline, search engine, data cleaning and querying, and, of course, your LLM and RAG. The main additions will be probably a graph database. You don't always have to use a graph database. You will maybe be using graphRAG. You'll have to set up how your RAG will interact with your graph. Graph specific operations for entity resolution, your fact verification and validation, and graph queries and APIs to and from your graph. Here you can see the typical ETL process from structured and unstructured data, but instead of populating a database table, you are populating the data model for your graph. During ETL, you will be running data validation, which will include validating that your data complies with your data model constraints. These can be SHACL or another form of data validation checking…

Contents