mayargamal97/Carbon-Literacy

Carbon Literacy Training in Kenya's Hospitality Sector # Carbon Literacy Training Analysis in Kenya's Hospitality Sector This repository contains data analysis and visualization work for assessing the impact of carbon literacy training in Kenya’s hospitality sector. The project aims to demonstrate how sustainability initiatives influence energy savings, carbon reduction, and revenue impact. ## Project Overview This analysis focuses on: - **Estimated Carbon Reduction** and **Energy Savings** due to training. - The **impact of pre- and post-training initiatives** on sustainability KPIs. - The **revenue impact** from eco-conscious practices. The **data cleaning and transformation** was performed using Python, while the **visualizations** were developed in Power BI. ## Repository Contents - **`carbon_literacy.py`**: Jupyter notebook containing the Python code for data cleaning and transformation. - **`carbon.png`**: Power BI dashboard visualizing the insights from the cleaned data. - **`final_dataframe.csv`**: Cleaned dataset in CSV format, ready for analysis and visualization. ## Data Cleaning Techniques The data cleaning process involved the following steps: 1. **Standardizing Text**: - Converted text in relevant columns (e.g., `KPI_Area`, `Global_Benchmark`, `Pre_Training`, `Post_Training`, `Estimated_Carbon_Reduction`, `Revenue_Impact`) to lowercase and removed any leading or trailing whitespace for consistency. 2. **Removing Parenthetical Information**: - Used regular expressions to remove any text within parentheses across several columns, keeping only the essential information. 3. **Extracting Percentage Values**: - Defined a custom function, `extract_percentage`, to capture and extract percentage values from text in the `Global_Benchmark`, `Post_Training`, and `Estimated_Carbon_Reduction` columns. - Extracted percentages were stored in new columns like `Post_Training_Percentage` and `Estimated_Carbon_Reduction_percentage`. 4. **Splitting Columns**: - Split the `Post_Training` column into separate columns …