Skip to content

Readme.md

Latest commit

 

History

History
155 lines (104 loc) · 3.64 KB

Readme.md

File metadata and controls

155 lines (104 loc) · 3.64 KB

🛠️ AI-Powered Support Ticket Classifier

A complete machine learning pipeline that classifies customer support tickets by issue type and urgency level, and extracts critical entities like product names, complaint keywords, and dates — all accessible through a sleek Gradio web interface.

📌 Project Overview

Customer support teams handle thousands of tickets daily. Manually classifying and extracting key information from these tickets is time-consuming and error-prone. This project automates:

  • 🔍 Ticket Classification: Predicts issue_type and urgency_level.

  • 🧠 Entity Extraction: Extracts key entities such as product names, complaint keywords, and dates.

  • 🌐 Gradio Interface: Allows end users to interactively classify tickets in a user-friendly UI.

🚀 Features

  • Data Cleaning & Preprocessing (lowercasing, lemmatization, stopwords removal)

  • TF-IDF & Feature Engineering (ticket length, sentiment scores)

  • Multi-Task Classification using Logistic Regression, SVM, and Random Forest

  • Entity Extraction using rule-based NLP (product names, dates, complaint terms)

  • Saved Models for inference reuse

  • Gradio Frontend for real-time predictions

📁 Project Structure

📦 ai_based_complaint_classifier
├── data/
│   ├── ai_dev_assignment_tickets_complex_1000.xlsx
│   └── cleaned_tickets.xlsx
│ 
├── models/
│   ├── issue_type_model.pkl
│   ├── urgency_level_model.pkl
│   ├── tfidf_vectorizer.pkl
│   └── feature_scaler.pkl
│ 
├── src/
│   ├── data_preprocessing.py
│   ├── feature_engineering.py
│   ├── model_training.py
│   ├── entity_extraction.py
│   └── predict_ticket.py
│ 
├── app/
│   └── app.py
│ 
├── .gitignore
├── LICENSE
├── requirements.txt
└── README.md

📦 Installation

1. Clone the repository:

git clone https://github.com/yashdew3/AI_Based-Complaint-Classifier.git
cd ai_based_complaint_classifier

2. Install the dependencies:

pip install -r requirements.txt

3. Run the Gradio App

cd src
python app.py

🧠 ML Models

  • Multi-class Classification

    • issue_type: Logistic Regression, SVM, Random Forest (best one selected)

    • urgency_level: Trained similarly

  • Entity Extraction

    • Regex + Keyword-based patterns for products, dates, and complaint terms

✨ Sample Input

Input:

"I received the EcoBreeze AC and it’s broken. Also, I ordered on April 5th."

Output:

{
  "issue_type": "Product Defect",
  "urgency_level": "High",
  "extracted_entities":
  {
    "products": ["EcoBreeze AC"],
    "dates": ["April 5th"],
    "complaint_keywords": ["broken"]
  }
}

📚 Dependencies

  • pandas, numpy, nltk, scikit-learn, textblob

  • gradio for UI

📌 Future Improvements

  • Integrate deep learning models (e.g., BERT)

  • Use Named Entity Recognition (NER) with spaCy or transformers

  • Support multilingual ticket classification

  • Database/API backend for enterprise usage

🤝 Contributing

Contributions are welcome! Please open an issue or submit a PR for enhancements or fixes. Feel free to check the issues page (if you have one) or open a new issue to discuss changes. Pull requests are also appreciated.

🧑‍💻 Author