Generate Research Ideas from PDFs using Infranodus, GraphRAG & Content Gap Analysis

This n8n workflow automates the process of extracting text from PDF documents, enriching it with GraphRAG (Graph-based Retrieval Augmented Generation) and Infranodus for content gap analysis, and then generating research ideas based on the combined insights. It simplifies the discovery of new research avenues and helps identify gaps in existing knowledge.

What it does

1. Triggers on Form Submission: The workflow starts when a user submits a form. This form is designed to accept a PDF file and potentially other parameters for the analysis.
2. Extracts Text from PDF: It takes the submitted PDF file and extracts its textual content. This step prepares the document for further processing.
3. Processes Text with GraphRAG (via HTTP Request): The extracted text is then sent to an external service (likely a GraphRAG endpoint) via an HTTP request. This service is expected to process the text, identify key entities, relationships, and concepts, and potentially generate initial insights or a knowledge graph representation.
4. Processes Text with Infranodus (via HTTP Request): Concurrently or sequentially, the text is also sent to another external service (likely an Infranodus API) via a separate HTTP request. Infranodus specializes in text network analysis and content gap identification, providing a different perspective on the document's content.
5. Combines and Analyzes Results (via Code Node): A Code node is used to combine and further analyze the outputs from both the GraphRAG and Infranodus services. This custom logic will synthesize the information, identify patterns, and pinpoint potential research ideas and content gaps.
6. Generates Research Ideas: Based on the combined analysis, the workflow generates a list of research ideas, potentially highlighting areas where more research is needed or where new connections can be made.
7. Outputs Results: The final research ideas and any relevant insights are presented as the output of the workflow.

Prerequisites/Requirements

• n8n Instance: A running n8n instance to host and execute the workflow.
• External GraphRAG Service: Access to an API endpoint for a GraphRAG (Graph-based Retrieval Augmented Generation) service. This service is crucial for extracting structured knowledge from the PDF content.
• External Infranodus Service: Access to an API endpoint for Infranodus. This service is used for text network analysis and content gap identification.
• API Keys/Authentication: Depending on your GraphRAG and Infranodus service configurations, you may need API keys or other authentication credentials configured in n8n's HTTP Request nodes.

Setup/Usage

1. Import the Workflow: Download the provided JSON and import it into your n8n instance.
2. Configure the "On Form Submission" Trigger:
   • Activate the workflow.
   • Note the webhook URL provided by the "On form submission" node. This URL will be used to submit your PDF files.
3. Configure the "HTTP Request" Nodes:
   • GraphRAG HTTP Request: Update the URL to point to your GraphRAG service's API endpoint. Configure any necessary headers (e.g., Content-Type, Authorization) and the request body to send the extracted PDF text.
   • Infranodus HTTP Request: Update the URL to point to your Infranodus service's API endpoint. Configure any necessary headers and the request body to send the extracted PDF text.
4. Customize the "Code" Node:
   • The "Code" node contains custom JavaScript logic to process the outputs of the GraphRAG and Infranodus services. You will likely need to modify this code to fit the exact structure of the data returned by your external services and to implement your specific content gap analysis and research idea generation logic.
5. Run the Workflow:
   • Once configured, you can test the workflow by submitting a PDF file to the webhook URL provided by the "On form submission" node.
   • Observe the execution in n8n to ensure all steps are running as expected and to review the generated research ideas.