📧 SMS/Email Spam Classifier

Welcome to the SMS/Email Spam Classifier project! This project is designed to help you identify whether a given SMS or email message is spam or not, using a machine learning model that I built and deployed as a Streamlit app.

🚀 Project Overview

1. Data Cleaning: The data was cleaned to ensure that only relevant and high-quality information was fed into the model.
2. Exploratory Data Analysis (EDA): In-depth analysis of the data was performed to understand patterns and distributions.
3. Text Preprocessing: Tokenization, stemming, and stopwords removal were applied to transform the raw text.
4. Model Building: Various models were tested, and the Multinomial Naive Bayes model achieved 100% precision.
5. Model Improvement: TF-IDF with a maximum of 3000 features was used to enhance accuracy.
6. Model Training: The model was trained and fine-tuned for optimal performance.
7. Streamlit App Development: A user-friendly web application was built using Streamlit.
8. Deployment: The app was deployed and is available for the community.

🎯 Features

• Real-time Classification: Instantly classify messages as "Spam" or "Not Spam".
• High Precision: The model is fine-tuned to achieve 100% precision.
• User-Friendly Interface: The Streamlit app is intuitive and easy to use.
• Community Access: The app is deployed and available for public use.

🔧 Installation

1. Clone the repository:

    git clone https://github.com/imAdnanSaid/sms-classifier.git
cd sms-classifier
  

2. Install the required packages:

    pip install -r requirements.txt
  

3. Run the Streamlit app:

    streamlit run app.py
  

🛠️ Technologies Used

• Python: The primary programming language.
• Pandas & NumPy: For data manipulation and analysis.
• Scikit-learn: For building and evaluating the machine learning model.
• NLTK: For text preprocessing.
• Streamlit: For building and deploying the web application.

📊 Model Performance

The Multinomial Naive Bayes model, improved with TF-IDF vectorization, achieved remarkable performance, particularly in precision, making it highly reliable for spam detection.

🌐 Deployment

The app is deployed on Streamlit and accessible here.

🤝 Contributing

Contributions are welcome! If you have suggestions for improvements or new features, feel free to open an issue or submit a pull request.

✨ Acknowledgments

- Special thanks to the Streamlit community for their support and resources.
- The datasets used for training the model were sourced from open data repositories.