This project aims to analyse the Gross Economic Data of 167 countries around the globe, to gain several insights. The analysis involves several steps, such as Data Cleaning, extensive visualisation, feature engineering and finally building the K-Prototype Clustering model.
Thus, the basic objectives pursued here are:
- Find trends in development between various features, and look into economic implications in such relationships.
- Perform unsupervised clustering based on all the features, and divide countries into clusters based on the features.
- Profile and understand the thus formed clusters, and their possible significance.
Here are some of the Visualisations (the complete project can be found in the colab notebook):
Count Based Distributions of various features:
Top and bottom countries of each Feature:
Pair-plots comparing each Feature:
As a result of clustering, three major clusters are observed. Taking a broader perspective, they are named based on help needed by the country in question.
Plots with clusters based on Categories:
World Map (Chloropleth):
