📖 Crypto Clustering - Unsupervised Machine Learning Project

📌 Project Overview

This project applies unsupervised machine learning techniques to analyze cryptocurrency market data. Using K-Means clustering, we group cryptocurrencies based on their 24-hour and 7-day price changes. Additionally, we optimize the clustering process by reducing dimensionality with Principal Component Analysis (PCA).

📁 Files in This Repository

📙 Crypto_Clustering.ipynb - The main Jupyter Notebook containing all the code and analysis.

📊 crypto_market_data.csv - The dataset used for clustering cryptocurrencies.

🗒️ README.md - This file, providing an overview of the project.

🛠 Technologies Used

🐍 Python

📙 Jupyter Notebook

🐼 Pandas

🤓 Scikit-learn (for K-Means & PCA)

📉 hvPlot (for interactive data visualization)

📊 Steps Completed

1️⃣ Data Preprocessing

Loaded the crypto_market_data.csv dataset.
Standardized numerical features using StandardScaler().

2️⃣ Finding the Best `k` (Elbow Method)

Applied K-Means clustering to the scaled dataset.
Used the Elbow Method to determine the optimal number of clusters.

3️⃣ Clustering Cryptocurrencies

Applied K-Means with the best k value to group cryptocurrencies.
Visualized the results using hvPlot scatter plots.

4️⃣ Dimensionality Reduction with PCA

Reduced the dataset to 3 principal components.
Calculated the total explained variance.

5️⃣ Finding the Best `k` Using PCA Data

Re-ran the Elbow Method on PCA-transformed data.
Determined whether k changed after PCA optimization.

6️⃣ Clustering with PCA Data

Applied K-Means clustering on the PCA-transformed data.
Visualized the new clusters.

7️⃣ Comparing Results

Compared original clustering vs. PCA clustering.
Analyzed the impact of reducing features with PCA.

📌 Key Findings

The best k value for clustering was found using the Elbow Method (4).
PCA helped reduce dimensionality while retaining most of the data variance.
The clustering results changed slightly after PCA.

📥 Installation & Setup

1️⃣ Clone this repository:

git clone https://github.com/Sharkb8t/CryptoClustering.git

3️⃣ Install required libraries (if not already installed):

pip install pandas scikit-learn hvplot

2️⃣ Open Jupyter Notebook Crypto_Clustering and run all cells sequentially.

Sharkb8t/CryptoClustering