📧 Spam vs Ham Email Classification using Machine Learning

This project builds a machine learning model to classify emails as spam or ham (legitimate) using Natural Language Processing (NLP) and Logistic Regression.

---

🔍 Problem Statement

Email spam detection is an essential task for filtering unwanted emails and improving user experience. This project aims to classify email messages into two categories:

• Ham (Not Spam) (label = 1)

• Spam (label = 0)

---

📁 Dataset

Source: Kaggle SMS Spam Collection Dataset

File Used: mail_data.csv

Features:

• Category: Spam or Ham

• Message: Email text

---

🛠️ Technologies Used

• Python (Google Colab)

• Libraries: NumPy, Pandas, scikit-learn

• Vectorization: TF-IDF (Term Frequency-Inverse Document Frequency)

• Modeling: Logistic Regression

---

⚙️ Data Preprocessing

Null Value Handling: Replaced null entries with empty strings

Label Encoding: Converted spam → 0 and ham → 1

Feature Extraction:

• Used TfidfVectorizer to transform email text into numerical feature vectors

• Data Splitting: Train-Test split (80% training, 20% testing)

---

🤖 Model Details

• Algorithm: Logistic Regression

• Input: TF-IDF features from email text

• Evaluation Metric: Accuracy Score

---

📊 Results

• Training Accuracy: ~96.8%

• Test Accuracy: ~95.0%

---

🧪 Usage

• Upload mail_data.csv to your Colab environment

• Run the notebook cells sequentially

• Enter a custom email message in the predictive system to check if it is spam or ham

---

🔮 Future Improvements

• Experiment with advanced models (SVM, Naive Bayes, or XGBoost)

• Use deep learning models (LSTM or Transformers) for better accuracy

• Add email metadata (sender, subject, etc.) as additional features