InfraNodus API: Convert Text to a Knowledge Graph with Network Analysis Insights

InfraNodus offers knowledge graph API access to its text network analysis and graph visualization algorithms. You can use the API to generate knowledge graphs from any text, extract main topical clusters, ideas, gaps, and network analysis insights. Our API can be used as a portable GraphRAG system to augment your existing LLM workflows.

This API can be especially useful for researchers, marketers, and SEO specialists who look to augment their insights with structural gap analysis and enhance their workflows by integrating the knowledge graph structure into their applications.

API Documentation
n8n No-Code Integrations
Make.Com No-Code Integrations

Examples of InfraNodus API Workflows with n8n

n8n is a popular tool for no-code integrations. It has an easy to use interface and a locally hosted version, which makes it easier to debug your workflows.

We created a set of n8n templates you can use to experiment with the insights that can be generated by our API.

Our most popular template helps you create an AI agent chatbot with expert knowledge that you can access directly in InfraNodus bypassing the need to set up a vector store in an external database. It uses GraphRAG under the hood, making the precision of responses much better than your standard RAG systems. GraphRAG will retrieve query-specific relations from the knowledge graphs you create with InfraNodus, unlike the standard RAG that will only perform similarity search. This is especially interesting for domain-specific applications and general / malformed user queries or where you need to add a reasoning logic to your LLM workflows.

Here is an example of the easiest implementation of GraphRAG with n8n:

As you can see, you can simply send a user's request to the API and let it generate the response for you — no need to set up OpenAI / agent nodes or complex vector store systems.

A more complex setup involves using our HTTP node (with the `requestMode=reprompt` parameter) to augment your prompt and to then send the improved user query to a knowledge base of your choice:

Finally, for more complex workflows, you can integrate our "expert" HTTP node to your AI agent workflow, where you use a single or multiple InfraNodus graphs as experts that can provide both reasoning logic and additional knowledge to your LLM workflows:

In this scenario, user's query is augmented using the InfraNodus GraphRAG reasoning expert, then sent back to OpenAI for an answer. The additional context is used to make the response less generic and introduce the reasoning logic contained in your knowledge graph. In our case, we use the ecological thinking framework knowledge graph as an expert:

You can access InfraNodus n8n workflow templates at:

• our GitHub repo n8n-infranodus-workflow-templates
(the most up-to-date workflows, constantly updated)

• the n8n website: n8n-infranodus-workflow-templates
(verified workflows with complete descriptions, updated every month by n8n team)

Examples of InfraNodus API Workflows (via Make.Com)

Make.Com is another no-code platform for integrations. It has a large library of templates and a user-friendly interface but lacks flexibilty of n8n and doesn't have a locally hosted version, which makes it more difficult to debug your workflows.

The templates prepared below will help you better understand possible workflows you can use InfraNodus for:

1. Question the news of the day and send ideas via Telegram

2. Generate topics and keywords for a collection of PDF documents

3. Generate research questions for scientific papers and articles

4. SEO: Reveal the current informational supply

5. SEO: Study the current informational demand

Some of the things you can do using the API:

• Get the main topical clusters (with AI-generated topic names and keywords included) for any discourse
• Identify structural gaps in a discourse: topics that are not so-well connected (for generating new ideas)
• Generate a text network graph for any text for network analysis (in Graph DOT and Graphology JSON formats)
• Generate research questions that bridge the structural gaps identified in a discourse
• AI-generated topical summaries (useful for generating better discourse summaries)

Data Privacy

You can set the `doNotSave` parameter to avoid saving graphs in your user account. This way your data won't even be stored in our logs and there will be no trace of what you process on our servers. We do not use any AI unless you explicitly specify that you need us to do that, in which case GPT models from OpenAI's API are used. OpenAI claims they don't use the data processed using their API for training purposes. Another advantage of using the knowledge graphs is that even if you end up using an AI, you will only send the graph structure to the model, not the underlying discourse, so you get an additional degree of privacy for your data.

Try the InfraNodus API