Smart-Ticket-Classifier 3.0

---

Overview

Smart-Ticket-Classifier is an AI-powered system that automatically categorises customer support tickets using Natural Language Processing (NLP) and Transformer models.
It helps support teams reduce manual triage, improve response times, and optimize workflows through intelligent automation.
Integrated with Slack for Real Time Communication and Backend Support.

Key Features

• Automated classification of support tickets
• Transformer-based NLP model for high accuracy
• Real-time or batch prediction support
• Interactive GUI dashboard
• Knowledge-base embeddings and analytics
• Shows Real time Solutions for the tickets
• Content gap detection and performance reports
• Analytics Part for the improvement
• Real time Screen Optimization
• Integrated Slack for Real time Backend support
• Multilingual supports 10+ different languages

---

What's New?

1. Analytics Part for the improvement

   

2. Real time Screen Optimization

   

3. Integrated Slack for Real time Backend support

   

---

⚙️ Installation

1. Clone the Repository

   git clone https://github.com/s4sahiko/Smart-Ticket-Classifier.git
   cd Smart-Ticket-Classifier
   

2. Create and Activate a Virtual Environment

   python3 -m venv venv
   source venv/bin/activate     # For Linux/Mac
   venv\Scripts\activate        # For Windows
   

3. Install Dependencies

   pip install -r requirements.txt
   

4. Prepare Dataset Use the provided cleaned_ticket_data.csv, Or replace it with your own dataset and update file paths in the scripts.

5. (Optional) Train the Model

   python trainer.py
   

6. Run the Application By running

   python app.py
   

7. Open the other terminal and Run to open the GUI Dashboard

   streamlit run gui_dashboard.py
   

Uplaod the ticket and Enjoy the saved time!

---

For Slack Integration

1. Paste the slack channel ID in app.py SLACK_CHANNEL_ID = "C0XXXXXXX"

2. Open terminal (in virtual env) paste your slack token (export SLACK_BOT_TOKEN=YOUR TOKEN)

3. Run app.py & gui_dashboard.py

---

Example Workflow

1. A new support ticket is received.

2. data_connector.py processes and cleans it.

3. categorizer.py uses the trained model to predict the category.

4. The result appears in the GUI or CLI With Real Time Solution Recommendation.

5. Analytics and reports are generated automatically.

6. Slack messaged can be seen from notification pannel.

---

Tech Stack

Language: Python

Libraries: Transformers, Pandas, NumPy, Matplotlib etc.

Frameworks: PyTorch

Visualization: Streamlit

Data: CSV-based datasets and embeddings

---

Why Use This Project

Efficiency — Automates ticket routing

Scalability — Handles large datasets easily

Accuracy — Uses Transformer-based contextual understanding

Insights — Detects knowledge gaps and tracks model performance

---

Contact & Support

For questions, bug reports, or suggestions — open an issue on the GitHub repository.

👤 Author: Rohith GRP:-4