n8n Workflow: Academic Citation Network Builder with PDF Vector API for Gephi Visualization

This n8n workflow is designed to process academic citation data, likely extracted from a PDF Vector API, and prepare it for visualization in Gephi or other network analysis tools. It focuses on transforming and structuring data to represent academic papers and their citation relationships.

What it does

This workflow performs the following key steps:

1. Initial Data Transformation (Edit Fields): This node is typically used to rename, remove, or modify fields in the incoming data. For academic citations, this might involve standardizing author names, publication years, or abstract fields.
2. Custom Code Execution (Code): This node allows for advanced data manipulation using JavaScript. In the context of academic citation networks, this could involve:
   • Parsing complex citation strings to extract individual references.
   • Creating unique identifiers for papers and authors.
   • Generating edge lists (source-target pairs) for citations.
   • Aggregating metadata for nodes (papers, authors).
   • Filtering out irrelevant data or handling missing values.
3. Output to Binary File (Write Binary File): The final processed data is written to a binary file. This is often used for exporting data in formats like CSV, JSON, or GEXF (for Gephi) that can be easily consumed by other applications.

Prerequisites/Requirements

• n8n Instance: A running n8n instance (self-hosted or cloud).
• Data Source: An upstream process or API that provides academic citation data. While not explicitly defined in this JSON, the directory name suggests a "PDF Vector API" is the intended source. The input to this workflow is expected to be the raw or semi-processed output from such an API.
• Understanding of Data Structures: Familiarity with how academic citation data is structured (e.g., authors, titles, publication venues, cited references) will be beneficial for configuring the "Edit Fields" and "Code" nodes.
• JavaScript Knowledge (for Code node): To customize the data transformation logic in the "Code" node.

Setup/Usage

1. Import the Workflow:
   • Download the provided JSON content.
   • In your n8n instance, go to "Workflows" and click "New".
   • Click the "Import from JSON" button and paste the workflow JSON.
2. Configure the "Edit Fields" Node (ID: 38):
   • Adjust the field transformations based on your input data structure to standardize or clean relevant fields (e.g., title, authors, year, abstract, references).
3. Configure the "Code" Node (ID: 834):
   • This is the most critical node for defining the network structure. You will need to write JavaScript code to:
     • Parse the input data to identify individual academic papers (nodes).
     • Extract cited references from each paper to establish connections (edges).
     • Format the output into a structure suitable for network visualization (e.g., a list of nodes with attributes and a list of edges with source/target IDs).
     • Consider generating data in a format like CSV (for nodes and edges separately) or GEXF/GraphML if supported by your Gephi import process.
4. Configure the "Write Binary File" Node (ID: 46):
   • Specify the File Name and File Path where the processed data should be saved.
   • Ensure the File Content is linked to the output of the "Code" node, formatted as a string or buffer representing your desired output file content (e.g., CSV string, JSON string).
   • Select the appropriate File Encoding if needed.
5. Trigger the Workflow:
   • This workflow does not have an explicit trigger defined in the JSON. You will need to manually execute it, or connect it to an upstream workflow (e.g., a "Webhook" or "Schedule" node) that provides the input data from your PDF Vector API.

Once executed, the workflow will process the input citation data and save the structured network data to the specified file, ready for import into Gephi for visualization.