This machine learning project was about clustering algorithm called K-means. It was written from scratch as an algorithmic exercise.
This project included:
- K-means clustering implementation
- Importing data from .CSV files
- Determinig the number of clusters (k)
- 4 different distance metrics
- Implementation of grouping quality measure - Silhouette
- Python (Pandas, NumPy, Matplotlib)
- Jupyter Notebook
Algorithm was tested on a few different datasets with diverse attribute numbers and arrangement of objects. All tests gave satisfactory results.
Labeled data:
Unlabeled data:
After clustering:
