This project segments customers using K-Means clustering and PCA (Principal Component Analysis). It generates synthetic customer data, runs clustering, and shows results as an interactive scatter plot using Plotly.
A classic machine learning project implemented as a Jupyter notebook, focusing on predicting passenger survival from the Titanic disaster. The notebook walks through the full data science workflow, including data loading, cleaning, and exploratory analysis of passenger demographics and ticket information.
Feature engineering and preprocessing steps are applied to transform raw data into model-ready inputs, followed by training and evaluating multiple machine learning models. Performance metrics and comparisons are used to assess predictive accuracy and understand model behavior.
The project serves as a clear, beginner-friendly example of supervised learning, demonstrating how structured tabular data can be used to build, evaluate, and interpret classification models in Python.