Linear Regression

This repository is a comprehensive introduction to linear regression, designed to help users understand one of the foundational algorithms in machine learning through practical examples and real-world applications.

Projects Overview

This repository contains 5 different linear regression projects:

1. Best Selling Video Games

• Data: Dataset CSV file in /Data/best-selling video games of all time.csv
• Notebook: Best Selling Video Games/Best Selling Video Game Prediction.ipynb
• Content: Linear Regression Model for Game Data with example Predictions.
• Objective: This project demonstrates the practical application of linear regression machine learning techniques to analyze and predict video game sales using real-world data from the best-selling video games of all time.
• Status: Ready for analysis

2. Energy Consumption

• Data: Energy consumption train and test dataset in Energy Consumption/Data
• Notebook: Energy Consumption/Energy Consumption.ipynb
• Content: Energy usage patterns and consumption data
• Objective: Predict energy consumption based on various factors
• Status: Ready for analysis

3. Formula 1

• Data: Multiple datasets in CSV format located Formula 1/Data
• Notebook: Formula 1/Formula 1 Analysis.ipynb
• Content: Formula 1 race results and performance metrics
• Objective: This analysis explores the critical relationship between qualifying performance, pit stop strategy, and final race outcomes using historical data from 1995 to the present.
• Status: Ready for analysis

4. Pokémon

• Data: Pokémon characteristics in Pokémon/Data/Pokemon.csv
• Notebook: Pokémon/Pokémon Identification.ipynb
• Content: Pokémon stats and characteristics
• Objective: Predict Pokémon stats based on other attributes
• Status: Ready for analysis

5. Retail Sales Trends

• Data: Sales data in Retail Sales Trends/Data/Warehouse_and_Retail_Sales.csv
• Notebook: Retail Sales Trends/Retail Sales Trends.ipynb
• Content: This dataset contains large-scale operational data from a retail and warehouse sales system. It includes sales volume, transfers, and supplier-product information.
• Objective: The goal is to predict the value of RETAIL SALES using machine learning and deep learning models.
• Status: Ready for analysis

Getting Started

1. Clone this repository
2. Install required Python packages for linear regression, note that each Notebook contains cell to install the dependencies, if required:
3. Choose a project folder and open the corresponding Jupyter notebook or in Google Collab
4. Follow the analysis and experiment with different regression techniques

Project Structure

Linear-Regression/
├── README.md
├── Best Selling Video Games/
│   ├── Best Selling Video Game Prediction.ipynb
│   └── Data/
│	├── best-selling video games of all time.csv
├── Energy Consumption/
│   ├── Energy Consumption.ipynb
│   └── Data/
│	├── test_energy_data.csv
│	├── train_energy_data.csv
├── Formula 1/
│   ├── Formula 1 Analysis.ipynb
│   └── Data/ (14 dataset CSV files)
├── Pokémon/
│   ├── Pokémon Identification.ipynb
│   └── Data/
│	├── Pokemon.csv
└── Retail Sales Trends/
│   ├── Retail Sales Trends.ipynb
│   └── Data/
│	├── Warehouse_and_Retail_Sales.csv    

Tips for Success

1. Data Exploration: Understand your data through visualization
2. Feature Engineering: Create meaningful features from raw data
3. Assumptions Check: Validate linear regression assumptions
4. Regularization: Use when dealing with overfitting
5. Cross-Validation: Evaluate model performance properly
6. Interpretation: Focus on understanding coefficient meanings

Last updated: 04/03/2026