Spam Text Classification

Overview

This project implements a spam text classification system using various Naive Bayes classifiers. It processes and analyzes text messages to determine whether they are spam or legitimate (ham).

Features

• Data preprocessing and feature extraction
• Implementation of multiple Naive Bayes classifiers:
  • Gaussian Naive Bayes
  • Multinomial Naive Bayes
  • Complement Naive Bayes
  • Bernoulli Naive Bayes
• Model training and evaluation
• Comparison of model performances
• Prediction functionality for new text inputs

Installation

To set up the environment, follow these steps:

1. Clone the repository:

  git clone https://github.com/yourusername/spam-text-classification.git

2. Navigate to the project directory:

  cd spam-text-classification

Usage

To run the classification system:

1. Ensure you have the dataset file '2cls_spam_text_cls.csv' in the project directory.
2. Run the Jupyter notebook or Python script:

  jupyter notebook Spam_Text_Classification_Naive_Bayes.ipynb

3. Follow the notebook cells or script execution to:

• Load and preprocess the data
• Train the models
• Evaluate model performance
• Make predictions on new text inputs

Model Performance

The project compares the performance of different Naive Bayes classifiers. Here's a sample of the results: