Spring2025_Scikit-learn

[Info-stats-ai]

Sure! Here's a version of the project description written in a more natural, first-person style—like you personally wrote it:

---

Project Title:

Time Series Analysis of Bitcoin Prices Using Scikit-learn

---

Project Description:

In this project, I explored how to use Scikit-learn to perform basic time series analysis on real-time Bitcoin price data. The main goal was to build a simple machine learning model that could predict future Bitcoin prices based on historical trends.

I started by fetching the data from the CoinGecko API, then cleaned and preprocessed it using pandas and Scikit-learn. This included handling any missing values, normalizing the data, and splitting it into training and testing sets.

After preparing the data, I implemented a linear regression model using Scikit-learn. I trained the model on the historical prices and then used it to predict future prices. To evaluate the model’s performance, I calculated the mean squared error (MSE) and visualized the results using Matplotlib, which helped me understand how well the model was tracking the actual price trends.

---

What I Used:

• Scikit-learn – for preprocessing and building the regression model
• Pandas – for handling and manipulating the time series data
• Matplotlib – for visualizing the results and trends
• Requests – to pull real-time price data from the CoinGecko API

---

Main Steps I Followed:

1. Got real-time Bitcoin price data using the CoinGecko API
2. Cleaned and preprocessed the data
3. Built and trained a linear regression model
4. Evaluated the model using MSE
5. Visualized the predictions vs. actual prices

---

What I Learned:

This project helped me understand how to work with time series data and use Scikit-learn for basic regression tasks. It also showed me the importance of good preprocessing and evaluation techniques in building reliable machine learning models.

---

Resources I Referred To:

• Scikit-learn documentation
• CoinGecko API docs
• Matplotlib tutorials