Apppublic

Harshu00718/worldbank-classifier2

sourceHugging Facecc-by-4.0updated 5mo agoView on Hugging Face

0likes

App README

🌍 World Bank Country Development Classifier

A Random Forest binary classifier that predicts whether a country is Developed or Developing based on World Bank international import trade data.

🎯 Model Performance

Metric	Score
Test Accuracy	98.84%
F1-Score	97.63%
ROC-AUC	99.92%
CV Mean (5-fold)	98.68% ± 0.13%

📦 Dataset

—Source: World Bank Import USD Enriched — Kaggle
—Records: 7,763 country-year records
—Countries: 217 | Years: 1974–2024
—Licence: CC BY 4.0

🤖 Model Details

—Algorithm: Random Forest (scikit-learn)
—Hyperparameters: nestimators=100, maxdepth=10, minsamplesleaf=5
—Features: Year, Log(Imports USD), Continent, Subregion, Income Group

🎓 ML Practical Assignment

Department of Computer Science & Engineering | Machine Learning Lab