Spam Mail Prediction using Machine Learning

A machine learning project that classifies emails as Spam or Ham (Not Spam) using TF–IDF vectorization and Logistic Regression. This project demonstrates a full ML pipeline including preprocessing, vectorization, model training, and evaluation.

Project Files SpamFilter.ipynb # Jupyter Notebook containing full implementation README.md # Project documentation Mail.csv # for email dataset

Features

Clean and simple ML pipeline

TF-IDF text vectorization

Logistic Regression classifier

High accuracy for spam detection

Easy to run on Jupyter Notebook or Google Colab

Technologies Used

Python 3

NumPy

Pandas

Scikit-learn

train_test_split

TfidfVectorizer

LogisticRegression

accuracy_score

Jupyter Notebook / Google Colab

Workflow

1. Import Dependencies

Basic ML and data preprocessing libraries.

2. Load Dataset

Emails containing text + labels (spam / ham).

3. Data Preprocessing

Clean text

Convert labels to numerical values

Handle missing data

4. Feature Extraction

Using TF-IDF Vectorizer:

TfidfVectorizer(min_df=1, stop_words='english')

5. Train–Test Split

Split dataset into 80% training and 20% testing.

6. Model Training model = LogisticRegression() model.fit(X_train, Y_train)

7. Evaluation

Accuracy measured using:

accuracy_score(Y_test, predictions)

Results

The model performs strongly for spam classification with high accuracy and generalization.

How to Run the Project Option 1: Google Colab (Recommended)

Upload the .ipynb file or open directly via Drive

Upload your dataset

Run all cells

Option 2: Local Machine pip install numpy pandas scikit-learn jupyter notebook

Future Enhancements

Add Naive Bayes, SVM, Random Forest models

Compare model accuracy

Build a web interface using Streamlit/Flask

Deploy as an API

Add real-time email classification UI

Author

Banoth Vikas Machine Learning & Software Engineering Enthusiast