Welcome to My Portfolio

Data Science & Machine Learning Enthusiast

About

About Me

Hello! I'm Ayan Mukherjee, a passionate data science enthusiast with a deep interest in turning complex data into actionable insights. My journey in data science started with a fascination for machine learning and artificial intelligence, where I quickly found myself drawn to the limitless possibilities these technologies offer.

When I’m not working on data-driven projects, I love exploring the world through different lenses. Photography allows me to capture the beauty around me, whether it's nature or candid moments. I also find joy in cooking, where I experiment with flavors and dishes that blend creativity with science. Playing the keyboard is another passion of mine, providing a peaceful escape through melodies. I also enjoy hitting the gym regularly, as it not only keeps me physically active but also fuels my mental clarity and discipline.

I’m constantly eager to learn and evolve, staying up to date with the latest trends and breakthroughs in the field of data science. Whether it's fine-tuning machine learning models or crafting a new recipe, I approach everything with curiosity and a desire to push my own boundaries.

Hand's on Experience

Project 1
Project 1

APS failure prediction uses sensor data to anticipate issues in a truck's air pressure system. By analyzing patterns in data, we can identify early warning signs of component failures, reducing downtime and costs. This predictive maintenance approach helps ensure vehicle safety and efficiency.

Source Code
Project 2
Project 2

This project developed a Random Forest model to accurately predict natural gas prices. The model analyzed historical data on various factors influencing prices. It was evaluated using metrics like MSE and RMSE, demonstrating its effectiveness in predicting future price trends. This project has practical implications for energy stakeholders.

Source Code
Project 3
Project 3

This project develops a deep learning model to classify lung x-ray images into disease categories (e.g., pneumonia, tuberculosis). The model uses a CNN architecture to extract features from the images and make accurate predictions, aiding radiologists in early disease detection and enabling remote healthcare.

Source Code

Skills

Python
Python
EDA
EDA
Feature Engineering
Feature Engineering
Machine Learning
Machine Learning

Contact Me