Academic Publication Data Extraction and Classification using AI and Google Sheets - n8n Workflow

Academic Researchers and Faculty needing automated tracking of publications.
University Administrators managing staff research output data.
Data Analysts who need to standardize unstructured HTML publication lists.
Users seeking powerful, customizable n8n templates for web scraping and AI processing.

Managing and cataloging academic publications from diverse faculty profile pages is a tedious, manual process. This expert n8n workflow solves this by providing a hands-free, scalable solution. It combines targeted web scraping with the intelligence of GPT-4 Mini to transform raw, messy HTML lists into clean, structured data.

This specific n8n workflow starts with a simple form submission, fetches the relevant webpage, and then uses a series of powerful n8n node components to extract, normalize, count, categorize, and store the data into a master Google Sheet and several categorized sheets (Journal, Conference, Book, etc.). This makes it an essential n8n template for any institution focused on research performance tracking.

The entire process is initiated by the On form submission n8n trigger. Once the Staff Name, Email, and URL are submitted:

1. Data Fetching: The workflow executes the Fetch website content n8n node, retrieving the HTML source of the specified publication URL.


2. Raw Extraction: The Extract all publications from the page n8n node isolates the publication list using defined CSS selectors, providing the raw text content.


3. AI Structuring: The Separate Each Publication node breaks the list into individual items. These items are then passed to the Generate Summary Report n8n node, which uses the GPT-4 Mini model (via the linked OpenAI Chat Model dependency) to reliably extract structured fields like authors, journalconference, and year based on a predefined JSON schema.


4. Master Storage & Notification: The structured data is simultaneously saved to the central 'Master Sheet' via a Google Sheets n8n node and converted into a CSV file by the Format as CSV Export node. The CSV is then sent as an attachment through the Send Notification Email Gmail n8n node to the submitting party.


5. Categorization & Aggregation: The Calculate Publication Statistics (Count) n8n node summarizes the data. The subsequent Switch n8n node uses sophisticated regex rules applied to the journalconference field to classify each publication type (e.g., 'journal', 'conference', 'book').


6. Sheet Routing: Based on the classification from the n8n node, the data is routed. It is sorted by year descending (using separate Sort nodes) and finally appended or updated in the corresponding category sheet (Journal, Conference, Book, Magazine, Patent, Other) using dedicated Google Sheets n8n nodes, completing this complex n8n workflow.

To deploy this powerful n8n workflow, follow these steps:

1. Import: Copy the provided JSON into your n8n instance using the 'Import Workflow' function.


2. Credentials Setup:

1. Trigger Activation: Activate the On form submission n8n trigger. You will use the generated form URL to input the staff profile links for processing.


2. Customization: Review the regex patterns in the Switch n8n node if your classification needs differ from the provided academic journal/conference patterns.

This n8n workflow relies on several specialized n8n node types to achieve its goal:

On form submission (Form Trigger): This serves as the primary n8n trigger, capturing the input parameters (URL, Staff Name, Email) needed to start the extraction process.
Fetch website content (HTTP Request): Performs the initial web request to scrape the target publication page URL defined in the n8n trigger data.
Extract all publications from the page (HTML): A critical web scraping n8n node. It uses the CSS selector .dropDownSummary.textArea ul ul li to precisely target the list items containing publication text.
Generate Summary Report (Information Extractor): This is the core AI n8n node, leveraging the gpt-4.1-mini model. It takes the raw publication text and forces it into a structured JSON format containing authors, journalconference, and year.
Save All to Master Sheet (Google Sheets): Appends the AI-extracted data to a single, comprehensive sheet, ensuring all results from this n8n workflow are centralized.
Calculate Publication Statistics (Summarize): Aggregates the data, counting occurrences based on unique combinations of journalconference and year before classification.
Switch (Logic): This highly configured n8n node handles routing. It uses extensive regular expressions to categorize the aggregated publications into distinct streams (Journal, Conference, Book, etc.), which is key to organizing the output data.
Send Notification Email (Gmail): Uses the extracted data, converted to CSV by the Format as CSV Export n8n node, to send a summary report back to the user who initiated the n8n trigger.