Email Classifier

Project Summery

An intelligent email classification system using Machine Learning and Natural Language Processing (NLP). This API can classify emails into categories like spam, personal, work, promotions, etc.

🚀 Features

• 🔍 Pre-trained ML model using Logistic Regression
• 🧠 Text preprocessing with TF-IDF
• ⚡ FastAPI backend with prediction endpoint
• 🧪 Swagger docs for testing API
• 🗂️ Easily extendable with more categories or models

🏗️ Project Structure

• email_classifier/
  • app/
    • main.py # FastAPI app for handling API routes
    • model.py # (Optional) Logic to load and predict using ML model
  • model/
    • email_classifier.pkl # Trained ML model file (TF-IDF + Logistic Regression)
  • train/
    • train_model.py # Script to train and save the model
    • dataset.csv # Email data with text and labels
  • requirements.txt # Python dependencies
  • README.md # Project documentation

🧠 Model

• Algorithm: Logistic Regression
• Vectorizer: TF-IDF
• Labels: spam, personal, work, promotions, updates, etc.

⚙️ Installation

1. Clone the repository:

   git clone https://github.com/sujit-codezen/ai_email_classifier.git
   cd email_classifier

2. Create virtual environment (optional but recommended):

   python -m venv env
   source env/bin/activate  # On Windows: env\Scripts\activate

3. Install dependencies:

   pip install -r requirements.txt

4. Train the Model:

   cd train
   python train_model.py

5. Run the API Server:

   cd ../app
   uvicorn main:app --reload

Output