Titanic Survival Prediction (3D Visualization)

Problem Statement

The Titanic dataset presents a classic machine learning challenge. The goal was to build a reliable model to predict passenger survival and, more importantly, to create an interactive platform that allows users to explore the complex factors influencing survival through intuitive 3D visualizations.

Technologies Used

Process & Challenges

The project followed an end-to-end machine learning pipeline, starting with data cleaning and feature engineering in Pandas. A Random Forest Classifier was trained for its high accuracy. The main challenge was translating the model's predictions into an interactive user experience. Streamlit was used to build and deploy the web app, allowing for real-time predictions. Creating meaningful and non-cluttered 3D scatter plots with Plotly to visualize relationships between age, fare, and passenger class was a key focus.

What I Learned

This project strengthened my skills in deploying machine learning models into production-ready web applications. I gained significant experience in building interactive dashboards with Streamlit and mastering advanced data visualization with Plotly to communicate complex data relationships effectively to a non-technical audience.