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

Unlock this course with a free trial

Join today to access over 25,300 courses taught by industry experts.

Automated fact verification

Automated fact verification

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

Start my 1-month free trial Buy for my team

Automated fact verification

“

- [Instructor] Validation from constraints helps you determine if your data complies, but that just means it conforms to what is expected based on your data model. It does not help pick a winner when there are disputes or verify the statement is actually correct or it's a legitimate dispute for that matter. Now, we will need to identify when there are statements in the graph or coming into the graph that do not agree. There are a lot of opinions, and disputed things in the world, so this is pretty common as we can see from our two water emission statements from earlier. And your LLM will need help deciding what to do with these disputes if they are important to your business. This usually happens as part of the fact verification pipeline in this middle part of the overall architecture you could use when using a graph with your LLM. So let's dive into the actual part of that diagram in the fact verification…

Contents