Code Clinic | Generating Knowledge Graphs with InstaGraph
An innovative application of GPT functions to force its response into a graph dataset
You might have seen this fast-growing repository on GitHub made by the creator of BabyAGI and one of the fastest VC brains in AI (Nakajima Yohei)
What does InstaGraph do? Well, it allows you to create knowledge graphs based on a URL or sentence. Knowledge Graphs? Sounds boring what is that?
A knowledge graph is a graph-structured data model that is made up of nodes, which represent entities, and edges, which represent relationships between entities.
TIL, Web 3.0 is on Wikipedia referred to as the Semantic Web. And I thought we would mean the crypto web. Guess you never stop learning. Anyways.
Knowledge graphs are useful for a variety of tasks, including:
Search and question answering: You can improve the accuracy and relevance of search results and question-answering systems. For example, Google's search engine uses a knowledge graph to provide rich snippets and knowledge panels for search results.
Recommendation systems: They can power recommendation systems that suggest products, content, or other items to users based on their interests and preferences. For example, Netflix uses a knowledge graph to recommend movies and TV shows to its users.
Data integration and analytics: Knowledge graphs can be used to integrate data from different sources and make it easier to analyze. For example, a knowledge graph could be used to integrate data from a customer relationship management (CRM) system and an enterprise resource planning (ERP) system to provide a unified view of customer data.
Again why is this interesting? I think this is interesting because InstaGraphs is…visual.
So, let me show it to you.
source (instagraph.ai)
But this is a code clinic and not a project review, so let’s dive in and install and play around with this interesting new tool and find out what it looks like under the hood.
If you haven’t done so, please like, share, and of course…subscribe.
If you are really really impatient. Here is the link to Yohei’s GitHub
One disclaimer before we start. Since this project is in active development, some code explainer segments might have changed and I am sure I will not update this post.