This is as an end-to-end data science project to support HELP International, a humanitarian NGO, in identifying countries that should be prioritized for aid allocation based on socio-economic and health indicators. The project addressed a real-world decision-making challenge: how to allocate limited humanitarian funding to countries with the greatest need. The analysis included data preprocessing, feature engineering, exploratory data analysis, standardization, dimensionality reduction using Principal Component Analysis, and clustering analysis using both K-means and hierarchical clustering. Key indicators such as child mortality, GDP per capita, income, health expenditure, life expectancy, inflation, imports, and exports were analyzed to group countries with similar development profiles. Clustering performance was evaluated using silhouette analysis, and each cluster was profiled to understand its socio-economic and health characteristics. A need-score ranking was created to identify the most vulnerable countries, with the final recommendation prioritizing countries with high child mortality, low GDP per capita, and low health expenditure. The project demonstrates how machine learning and statistical analysis can support data-driven humanitarian aid allocation and decision-making. Technologies used include R, R Markdown, tidyverse, ggplot2, PCA, K-means clustering, hierarchical clustering, silhouette analysis, and data visualization.