Below is a sample notebook on my kaggle dashboard which showcases how clustering was implemented to group the states in the US using different crime numbers.
Various aspects of the notebook:
- Data normalization
- Fit K-means model
- Get scree plot and decide on the number of clusters
- Fit final K-means model and get clusters
- Write clusters column along with others into a CSV
- Visualize clusters for different crimes on a scatter plot and among states on tableau