This project analyzes global life expectancy trends from 1960–2022, focusing on gender-based differences and country-level insights. It combines Python for data wrangling and machine learning, MySQL for structured data storage, and Power BI for interactive dashboarding.
- Trend analysis, gender gap visuals, top countries, and correlation heatmaps
- Random Forest Regression
- (R² ≈ 0.98, MAE ~1.0) to predict life expectancy by gender and year
- ARIMA models for 5–10 year forecasts by country, including top and bottom gender gap nations
- Predicted vs actual plots, country-wise life expectancy trends, gender gap charts, and global ranking bars
- Created a relational database
life_expectancy_dbusing SQLAlchemy - Stored modeled outputs as a table
- Configured user access and permissions for secure connectivity from BI tools
- Connected directly to MySQL for live data visualization
- Built a multi-page interactive dashboard with:
- Scatter plot of male vs female life expectancy by country
- Stacked area chart showing trends over time
- Bar chart for Top 10 countries with highest gender gap
- Treemap and KPI card to highlight average gender gap by year
- Enabled filters for Year and Country to explore data dynamically
- Source: Cleaned from World Bank life expectancy datasets (separated by Male and Female)
- Shape: ~17,000 rows per gender, covering 200+ countries over 60+ years
- Format: Long-format with columns for Country, Country Code, Year, and Life Expectancy
Links to dataset:
Life expectancy at birth, male (years)- https://data.worldbank.org/indicator/SP.DYN.LE00.MA.IN?end=2022&start=2013
Life expectancy at birth, female (years) - https://data.worldbank.org/indicator/SP.DYN.LE00.FE.IN