Data Analytics Projects

Advanced analytics solutions leveraging machine learning, statistical modeling, and data visualization to drive business decisions

Featured

Classification Modeling

E-Commerce Customer Churn Prediction

Developed classification models (Decision Tree, Logistic Regression) achieving 89.3% accuracy to predict customer churn. Identified key drivers including early tenure (0-3 months), customer complaints (~80% churn rate), and cashback patterns. Analysis suggests targeted retention strategies could reduce churn from 16.8% to below 12%.

89.3% Accuracy

29% Churn Reduction

52% Churner Recall

Python scikit-learn Logistic Regression pandas numpy Stratified sampling

Code Demo

Featured

Predictive Analytics

Predictive Customer Lifetime Value (CLV) Modeling

Built machine learning models achieving 98% accuracy to forecast customer value, enabling data-driven marketing budget allocation and retention strategies. Identified top 20% customers driving 45% of revenue through RFM segmentation, projected to increase CLV by 20% and reduce high-value churn by 25%.

98% Model Accuracy

20% CLV Increase

25% Churn Reduction

Python Scikit-learn XGBoost Random Forest Gradient Boosting RFM Analysis

Code Demo

Python Random Forest Statistical Analysis

Walmart Sales Forecasting & Optimization

Optimized sales strategies using predictive models and statistical analysis to forecast weekly sales patterns. Identified key drivers including temperature, fuel prices, and holiday effects, enabling data-driven inventory management and promotional planning for improved revenue performance.

Python Scikit-learn Random Forest Statistical Analysis Matplotlib Seaborn

Code Demo

Python Plotly Data Visualization

US Air Quality Analysis

Comprehensive analysis of U.S. air pollution trends using AQI data to compare regional patterns, track temporal changes, and assess environmental health impacts. Created interactive visualizations revealing critical insights for policy recommendations and public health initiatives.

Python Pandas NumPy Matplotlib Seaborn Plotly

Code Demo

Tableau Geospatial Analysis

Global Forest Watch Analysis

Analyzed global deforestation patterns, carbon emissions, and forest cover changes using satellite data. Created comprehensive Tableau dashboards visualizing regional trends, enabling stakeholders to identify high-risk areas and prioritize conservation efforts effectively.

Tableau Excel Data Cleaning Geospatial Analysis Dashboard Design

Code Demo