Apply Clustering Algorithm for Recommendation System

**Describe the solution you'd like**
Apply K-Means Clustering or Gaussian Mixture Model for Recommendation System. You can either use One Hot Encoding or Label Encoding.

No file, test, or entry point is named in the issue; first locate the existing recommendation-system notebook and inspect its data-cleaning and EDA flow. Compare the proposed K-Means and Gaussian Mixture approaches with the available encoded data, and define completion as an integrated clustering-based recommendation result with a clear evaluation.

Written by the indexing model from the issue text.