This project performs an end-to-end exploratory data analysis (EDA) on global layoffs data using Python. It aims to uncover patterns, trends, and insights related to workforce reductions across industries, companies, and regions.
To analyze layoffs data and identify:
- Industry-wise impact
- Company-level trends
- Geographic distribution
- Time-based patterns
- Key influencing factors
The dataset includes information such as:
- Company name
- Industry and location
- Total employees laid off
- Percentage of layoffs
- Funding raised
- Company stage
- Date of layoffs
- Converted date column to datetime format
- Removed duplicate entries
- Standardized categorical columns (industry, country)
- Handled missing values in numerical and categorical fields
-
Extracted year and month from the date column
-
Created impact categories based on layoff percentage
- Low
- Medium
- High
- Critical
The following analyses were performed:
- 🔹 Top companies by layoffs
- 🔹 Industry-wise layoffs distribution
- 🔹 Country-wise layoffs (global map)
- 🔹 Monthly and yearly layoff trends
- 🔹 Correlation analysis between variables
- Bar charts (companies & industries)
- Line plots (time-series trends)
- Choropleth map (global layoffs distribution)
- Heatmap (correlation matrix)
- Layoffs are concentrated in specific industries such as consumer and retail
- Certain countries show higher layoffs due to strong tech ecosystems
- Layoffs occur in waves rather than uniformly over time
- Weak correlation between funding and layoffs indicates multiple influencing factors
- Python
- Pandas, NumPy
- Matplotlib, Seaborn
- Plotly
- Google Colab
This project highlights how data analysis can be used to uncover meaningful insights from real-world datasets. It demonstrates the importance of data cleaning, feature engineering, and visualization in understanding complex trends like layoffs.
- Build an interactive dashboard using Streamlit
- Apply machine learning models to predict layoffs
- Integrate real-time data sources
Kamakshi Pal Aspiring Data Scientist