This repository contains code and analysis for a set of data science projects focused on modeling, visualization, and signal processing.
I write about the projects on Medium, and keep everything organized on dbolotov.github.io.
- Predicting Gym Member Churn to Support Targeted Retention: Using behavioral and lifecycle data to reduce early membership loss. Jupyter notebook | Medium article
- The Thermal Decay of Beverages: A Casual Overanalysis 🍵: Real data and Newton’s Law of Cooling to compare how mugs lose heat over time. Jupyter notebook | Medium article
- Time Series Smoothing: A comparison of six different smoothing techniques on noisy data. Jupyter notebook | Medium article | Interactive Streamlit app
- Severe Weather in California (2000–2024): An exploratory analysis of NOAA weather event data, with plots of damage, injuries, and deaths across event types and time. Jupyter notebook (nbviewer)