Portfolio Details

Each project in this portfolio reflects my hands-on experience in solving real-world problems using data science, machine learning, and full-stack development. From predictive models to intelligent web applications, these solutions are built with a focus on scalability, functionality, and impact.

Project Information

  • Project Name: Email Spam Classifier
  • Category: Text Classification / NLP
  • Model Type: Multinomial Naive Bayes
  • Github Repo: Link

Technologies Used

This project is built using Python with Scikit-learn and Natural Language Processing (NLP) techniques. It uses TF-IDF vectorization and a Multinomial Naive Bayes model to classify emails as spam or not spam. The application also includes data preprocessing steps like text normalization, stopword removal, and tokenization. Visualization libraries like Matplotlib and Seaborn were used for EDA.

Details of Project

The Email Spam Classifier is an intelligent NLP solution designed to automatically identify and filter spam messages. It processes raw email text using TF-IDF vectorization and applies a Multinomial Naive Bayes model to predict whether an email is legitimate or spam.

The system is trained on a labeled dataset of thousands of messages, achieving high accuracy with efficient computation. It's optimized to handle new, unseen data with minimal latency and can be integrated into email platforms or security pipelines for real-time filtering.

Project Features

  • Classifies emails as "Spam" or "Not Spam" using NLP techniques.
  • Preprocessing pipeline: stopword removal, lowercasing, tokenization.
  • TF-IDF vectorizer and Naive Bayes model for fast, reliable predictions.
  • Evaluation through confusion matrix, accuracy score, and precision-recall metrics.
  • Lightweight and ready for integration into any Flask or FastAPI service.