E-Mail_Spam_Detector_using_Machine_Learning

This project demonstrates a machine learning approach to classify emails as spam or not spam using a Naive Bayes classifier. It leverages Natural Language Processing (NLP) techniques to convert email text into numerical features, making it a straightforward and practical application for beginner to intermediate Data Scientists.

Project Goals

• Objective: To build a model that classifies emails as either spam or not spam based on their content.
• Dataset: A sample dataset of 500 emails was created with random spam and non-spam email content.
• Target Audience: This project is intended for senior Data Scientist employers or anyone looking to develop skills in NLP and binary classification.

Project Overview

This project walks through:

1. Data Preparation: Creation of a dataset containing 500 emails labeled as spam or not spam.
2. Data Processing: Using Count Vectorizer to transform text data into numerical form.
3. Modeling: Training a Naive Bayes classifier for binary classification.
4. Evaluation: Measuring model performance with accuracy and classification reports.

File Structure

• email_spam_detector.ipynb: Jupyter Notebook containing all code, explanations, and output.
• data: Contains sample email data used in this project.
• README.md: Project documentation.

Getting Started

To run this project on Google Colab or locally, ensure you have the required libraries installed.

Prerequisites

• Python 3.x
• Libraries:

  pip install pandas numpy scikit-learn

Running the Project

1. Clone this repository or download the email_spam_detector.ipynb file.
2. Open the notebook in Google Colab or a Jupyter Notebook environment.
3. Run each cell to process data, train the model, and view results.

Dataset

This project uses a sample dataset with 500 rows of randomly generated email texts, divided evenly between spam and non-spam labels. Below is an example of the dataset structure:

Email Text	Label
"Congratulations! You've won a $1,000 gift card. Click here to claim now!"	spam
"Meeting is scheduled at 3 PM tomorrow, please confirm."	not spam
"Last chance to win a trip to Hawaii!"	spam
"Hello, wanted to check if you're available for a quick call."	not spam

The emails cover various patterns found in spam and non-spam messages, providing the model with basic yet distinct training data.

Code Walkthrough

Step 1: Data Preparation

• We create a dataset of 500 emails, labeled as either spam or not spam.
• Labels are assigned randomly, with equal distribution for model balance.

Step 2: Data Processing

• Using CountVectorizer to convert email text into numerical features suitable for Naive Bayes classification.

Step 3: Model Training

• We use a Naive Bayes classifier due to its efficiency and effectiveness for text classification.

Step 4: Evaluation

• The model is evaluated using accuracy and a classification report to assess precision, recall, and F1-score.

Results

• Accuracy: Achieved around 80-85% accuracy on the test dataset.
• Classification Report: Shows the precision, recall, and F1-score for each class.

Sample Code

Here's a snippet showing the model training and evaluation process:

from sklearn.naive_bayes import MultinomialNB
from sklearn.metrics import accuracy_score, classification_report

# Model Training
model = MultinomialNB()
model.fit(X_train, y_train)

# Prediction and Evaluation
y_pred = model.predict(X_test)
print("Accuracy:", accuracy_score(y_test, y_pred))
print("\nClassification Report:\n", classification_report(y_test, y_pred))

Future Improvements

• Expand Dataset: Increase the dataset size with more diverse email texts.
• Advanced Models: Experiment with other algorithms, such as Support Vector Machines or Random Forest.
• Feature Engineering: Apply techniques like TF-IDF for more nuanced text feature representation.

License

This project is open-source.

Contributions

Feel free to fork this repository, submit issues, and open pull requests to enhance the project.

Acknowledgments

Thanks to the open-source community and various resources that made this project possible. This project is designed to be a practical example for those looking to enter the field of NLP and binary classification with machine learning.