How aswina1636/global_cost_of_living_analysis-using-world_bank-pli_data-1990-2024 works
This project turns World Bank Price Level Index data into cost-of-living analysis, visualizations, and country-category predictions. The user-facing workflow is the Gradio dashboard: users explore PLI by year and country or submit PLI and year for a classification. The README also documents preprocessing, feature engineering, EDA, and a Power BI dashboard. Notebook roles and the exact wiring between artifacts are not shown in the excerpts, so those relationships are limited to README-documented workflow links.
Main entry points: ML_Pipeline.ipynb PLI Features · app.py Random Forest · cleaned_data.csv Cleaned Data · raw_data_of_world_bank_PLI.csv Raw PLI Data
Components 14
Data Preparation
- World Bank PLI
raw_data/raw_data_of_world_bank_PLI.csv - Raw PLI Data
raw_data/raw_data_of_world_bank_PLI.csv - Cleaned Data
cleaned_data.csv - Filtered Data
filtered_data.csv - Data Workflow
main.ipynb
Economic Analysis
- Exploratory Analysis
ML_Pipeline.ipynb - PLI Features
ML_Pipeline.ipynb
Classification
- Random Forest
app.py - Saved Model
model.pkl - RF Model Artifact
rf_model.pkl - Neural Network Work
neural_network.ipynb
User Products
- Gradio Dashboard
app.py - Power BI Dashboard
dashboard.pbix
Other components
- Analyst
Connections 14
- World Bank PLI → Raw PLI Data: supplies
README.md - Raw PLI Data → Data Workflow: preprocesses
README.md - Data Workflow → Cleaned Data: produces
README.md - Cleaned Data → Filtered Data: filters
- Cleaned Data → PLI Features: engineers
README.md - PLI Features → Exploratory Analysis: analyzes
README.md - Analyst → Gradio Dashboard: explores
README.md - Gradio Dashboard → Random Forest: invokes
app.py - Random Forest → Saved Model: uses
- PLI Features → Random Forest: trains
README.md - Analyst → Power BI Dashboard: explores
README.md - PLI Features → Neural Network Work: experiments
- Exploratory Analysis → Power BI Dashboard: informs
- RF Model Artifact → Random Forest: persists