🎫 Classify sentiment of incoming text using Hugging Face, Google Sheets, and Jira

⚡ 18 views · 🎫 Ticket Management & Triage

Description

Sentiment Analysis Workflow using Webhook, Hugging Face, Google Sheets & Jira

This workflow automatically analyzes incoming text feedback, classifies it into Positive, Neutral or Negative using a Hugging Face sentiment model, stores results in Google Sheets and creates Jira tickets for negative feedback.

Quick Steps to Get Started

Import the workflow into your n8n account
Set up Webhook endpoint (/sentiment-input)
Add Hugging Face API token in HTTP Request node
Configure Google Sheets (3 tabs: Positive, Neutral, Negative)
Connect Jira credentials
Activate the workflow
Send POST request with text data

What It Does

This workflow automates sentiment analysis of incoming text data using a machine learning model hosted on Hugging Face. It receives multiple text inputs via a webhook, processes each input individually and evaluates sentiment scores returned by the model.

The workflow intelligently determines whether the sentiment is Positive, Neutral or Negative based on the highest score and a confidence threshold. It ensures more reliable classification by applying a score validation logic.

Once classified, the workflow routes the data accordingly. Each sentiment category is stored in a separate Google Sheets tab, making it easy to track and analyze feedback trends. Additionally, any negative feedback automatically triggers the creation of a Jira ticket, enabling quick issue resolution.

Who It’s For

Businesses collecting customer feedback
Product teams monitoring user sentiment
Customer support teams
SaaS companies handling reviews or complaints
Developers building AI-powered automation workflows

Requirements

To use this workflow, you need:

n8n account (self-hosted or cloud)
Hugging Face API access token
Google Sheets account with service account credentials
Jira Software Cloud account
Basic understanding of n8n workflows and nodes

How It Works & Setup Guide

Step 1: Import Workflow

Import the JSON file into n8n

Step 2: Configure Webhook

Node: Receive Feedback
Method: POST
Endpoint: /sentiment-input
Input format:

{
  "data": [
    { "text": "I love this" },
    { "text": "This is okay" },
    { "text": "Worst experience ever" }
  ]
}

Step 3: Split Input Data

Node: Split Text Items
Splits array into individual items for processing

Step 4: Rate Limiting

Node: Rate Limit Control
Prevents API overload (optional delay control)

Step 5: Preserve Input

Node: Preserve Input Text
Keeps original text intact for later use

Step 6: Sentiment Analysis

Node: Get Sentiment Scores
Add Hugging Face API token in headers:
- Authorization: Bearer YOUR_API_KEY
Model used: cardiffnlp/twitter-roberta-base-sentiment

Step 7: Merge Data

Node: Merge Text & Scores
Combines model output with original input

Step 8: Compute Sentiment

Node: Compute Sentiment
Logic:
- Highest score determines sentiment
- Confidence threshold: > 0.9
- Otherwise defaults to Neutral

Step 9: Route Data

Node: Route by Sentiment
Routes into:
- Positive
- Neutral
- Negative

Step 10: Store Results

Nodes:
- Store Positive Feedback
- Store Neutral Feedback
- Store Negative Feedback
Append data into respective Google Sheets tabs

Step 11: Create Jira Ticket

Node: Create Jira Ticket
Triggered only for Negative sentiment
Includes:
- Text
- Sentiment
- Timestamp
- Priority: High

How To Customize Nodes

Webhook Node
- Change endpoint path as needed
HTTP Request Node
- Replace Hugging Face model with another model if required
Compute Sentiment Node
- Adjust confidence threshold (currently 0.9)
- Modify classification logic
Google Sheets Nodes
- Change document ID or sheet names
- Add more columns if needed
Jira Node
- Customize issue type, priority or project

Add-ons

You can enhance this workflow with:

Email or Slack notifications for negative feedback
Dashboard visualization using BI tools
Sentiment trend analytics
Multi-language sentiment analysis
Integration with CRM systems

Use Case Examples

Customer feedback analysis for apps or websites
Product review classification from multiple sources
Social media sentiment monitoring
Support ticket prioritization system
Survey response automation

There can be many more use cases depending on how feedback data is collected and used.

Troubleshooting Guide

Issue	Possible Cause	Solution
Webhook not triggering	Incorrect endpoint or method	Verify POST request and URL
No sentiment output	API token missing/invalid	Check Hugging Face API key
Incorrect sentiment classification	Confidence threshold too high	Adjust threshold in Compute node
Data not appearing in Sheets	गलत credentials or sheet mapping	Reconnect Google Sheets
Jira ticket not created	Jira credentials issue	Verify Jira API connection

Need Help?

If you need assistance setting up this workflow, customizing nodes or building advanced automation solutions, feel free to reach out to n8n developers at WeblineIndia.

Our team can help you:

Deploy n8n workflows on cloud/server
Customize AI-based automation
Integrate APIs and enterprise tools
Build scalable business automation systems

Contact WeblineIndia for expert support and tailored solutions.

🔗 Nodes Used

Google Sheets, HTTP Request, Webhook, Jira Software

📥 Import

Download workflow.json and import into n8n: Workflow menu → Import from File

📖 Importing guide · 🔑 Credential setup