📧 Spam detection application for detecting unwanted e-mails getting to a user's inbox

📌 Project Overview

The Email Spam Detection Application is designed to identify and filter out unwanted emails, ensuring that users' inboxes remain free from spam. The project leverages Machine Learning (ML) to classify emails as either Spam or Ham (Not Spam). The application consists of:

• A Machine Learning Model trained to detect spam emails.
• A Flask Backend to serve the model and handle classification requests.
• A Web Frontend built with HTML, CSS, and JavaScript to allow users to input emails for spam detection.

🚀 Features

• Email Spam Classification: Users can input an email message, and the system will classify it as Spam or Not Spam.
• ML Model Integration: A trained model packaged using joblib.
• Web-Based Interface: A simple and user-friendly interface for email classification.
• RESTful API: The Flask backend provides an API endpoint for classification requests.

---

🛠️ Project Development Process

📊 Data Collection

• The dataset for training the model was collected by Ebenezar Noble and Adelakun Joshua.
• The dataset includes labeled email samples indicating whether they are spam or ham.
• The data was preprocessed to remove noise, stop words, and unnecessary elements.

🤖 Model Development

1. Preprocessing & Feature Engineering

   • Tokenization, removal of special characters, and conversion to lowercase.
   • Stopword removal and stemming/lemmatization.
   • Text vectorization using TF-IDF (Term Frequency-Inverse Document Frequency).

2. Model Selection & Training

   • Tried different models including Naïve Bayes, Logistic Regression, and SVM.
   • The best performance was achieved using Multinomial Naïve Bayes, which is ideal for text classification.
   • The model was trained on the dataset and evaluated using accuracy, precision, recall, and F1-score.

3. Model Packaging & Deployment

   • The trained model was serialized using joblib.
   • Integrated the model into a Flask API.

🌐 Web Application Development

🏗️ Backend Development (Flask API)

• Developed by: Azuka Ifeanyichukwu (Lead, Backend & Model Training) with assistance from David Johnson.
• A Flask REST API was built to:
  • Load the trained model.
  • Accept user input via an API endpoint.
  • Process input text and return the classification result.

🎨 Frontend Development

• Developed by: Adeoye Ayonifeade, Oni Davies, and Olusanya Joseph.
• Built using HTML, CSS, and JavaScript.
• Designed a user-friendly UI where users can:
  • Input an email.
  • Click a button to classify the email.
  • Receive real-time spam detection feedback.

🔬 Testing & Debugging

• Tested by: Ayonifeade and David Johnson.
• Conducted unit testing on:
  • Model performance using test datasets.
  • API response times and accuracy.
  • Frontend usability and integration with the backend.

📂 Deployment & Version Control

• GitHub Repository Management by Azuka Ifeanyichukwu.
• Code was pushed to GitHub for collaboration and version control.

---

👥 Team Contributions

Name	Role
Adelakun Joshua(210202001)	Data Collection and Project Manager
Ebenezar Noble(200202018)	Data Collection
Azuka Ifeanyichukwu(220210014)	Model Training, Flask Backend, GitHub Repo Management
David Johnson(220202043)	Assisted Backend, Testing
Adeoye Ayonifeade(210202003)	Frontend Development, Testing
Oni Davies(210202027)	Frontend Development
Olusanya Joseph(210202025)	Frontend Development

---

🏁 Running the Application

🔹 Prerequisites

Ensure you have the following installed:

• Python 3
• Flask
• Joblib
• Scikit-learn

🔹 Steps to Run Locally

1. Clone the Repository

   git clone https://github.com/oladapokayodeabiodun/Project3.git
   cd spam-detection-app

2. Set Up Virtual Environment

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

3. Install Dependencies

   pip install -r requirements.txt

4. Run the Flask Backend

   python app.py

5. Access the Frontend
   • Open index.html in a browser.